<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Description
This pull request introduces a complete Russian translation for ImHex,
ensuring that Russian-speaking users can navigate and utilize the
software in their native language. The translation covers all user
interface elements, menus, dialogs, and messages, providing a seamless
experience for Russian users.
### Testing
The translation has been thoroughly tested to ensure accuracy and
completeness. All translated text has been reviewed for grammatical
correctness and contextual appropriateness.
### Screenshots
Co-authored-by: Lemon4ksan <senya20151718@gmail.com>
Fixes#2084
Error in text editor prevented using shortcuts when language other than
English was chosen.
The code was mistakenly using localized name to test which window had
focus a execute the shortcut.
Fixed it by switching the name of the child window to the constant value
used to check the windows focused identity.
### Problem description
This PR addresses issue #2013 that described a cluttered Event Manager.
This is a DX issue and should not impact the users whatsoever.
### Implementation description
The changes revolve around three main points:
1. the Event Manager (`event_manager.hpp`) was split into four
categories: GUI, Interaction, Lifecycle, and Provider, and two types:
Events, and Requests. This results in the following files:
- `events_gui.hpp`
- `events_interaction.hpp`
- `events_lifecycle.hpp`
- `events_provider.hpp`
- `requests_gui.hpp`
- `requests_interaction.hpp`
- `requests_lifecycle.hpp`
- `requests_provider.hpp`
2. Every event and request now has its own piece of documentation, with
a `@brief`, accompanied by a longer comment if needed, and gets its
`@param`s described.
3. The old `event_manager.hpp` import was removed and replaced by the
correct imports wherever needed, as to reduce spread of those files only
to where they are truly useful.
### Additional things
The commits have been split into (chrono-)logical steps:
- `feat`: split the Event Manager, and replace the imports
- `refactor`, `chore`: make various small changes to match the required
structure
- `docs`: add documentation for events and requests
Hopefully, this will help to review the PR.
*Note: Beware of very long rebuild times in between the commits, use
them sparingly! The Actions will ensure this PR builds anyways*
Closes#2013
---------
Signed-off-by: BioTheWolff <47079795+BioTheWolff@users.noreply.github.com>
Co-authored-by: Nik <werwolv98@gmail.com>
The previous implementation seems to have been broken by Imgui updates.
There is also some improvement in focus handling and also a bug in
replace where the last match was not being replaced has been fixed.
Fixed also slowdown in large files when only one char was typed by not
searching until enter has been pressed.
Added key repetitions for enter and arrows to be able to handle large
number of matches and fixed some formatting problems.
The recently added data checks allowed invalid sized vertex arrays to be
used as if they were valid making ImHex crash.
Moved all the error messages into localization strings for translation.
After successfully finding matches and setting the cursor to them, the
screen would jump to the original window location upon closing the
window.
The error was caused by the wrong assumption that the scroll location
should be restored when window is closed. Instead, the right amount of
scrolling needs to be calculated to account for the window no longer
covering part of the text editor. Unused variable was discarded.
Another unrelated error is that the history of search names cannot be
accessed which will be addressed at a later PR.
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
See #2033 (`hex::dec::lzma_decompress` reports an error when
decompressing a small buffer).
### Implementation description
<!-- Explain what you did to correct the problem -->
Set the LZMA decompressor memory limit to 1GiB fixed. Print a warning
when exceeded, and abort with returning `false`.
### Screenshots
<!-- If your change is visual, take a screenshot showing it. Ideally,
make before/after sceenshots -->
Normal result when decompressing a small buffer
Warning message when `memlimit` is exceeded: (Set to 64B here for a
demo)
> W: lzma_decompress memory usage 1114168 bytes would exceed the limit
(64 bytes), aborting
### Additional things
<!-- Anything else you would like to say -->
Is the warning wording OK? I'm not a native English speaker so please
change it if you want to.
Some issues related to the padding added to scroll past the end for
console that has padding added.
Added a shortcut to scroll editors one pixel at a time.
Fixed whole lines always drawn at the top even if scroll value is chosen
so that only a portion of the top line is visible. This caused errors in
horizontal scrolling.
Fixed Ctrl-F Ctrl-G and Ctrl-H messing the editor display.
Fixed the end of the line could not be clicked with mouse
Fixed line numbers and their lines could be displayed at different
heights.
Made numbers that represented lines floats instead of integers to allow
partial line display.
Two major improvements:
1) see through scrollbars when not hovered.
2) un-scrollable line numbers.
Also enlarged display region by eliminating padding. There is still a
problem with lines jumping when the scrollbar is dragged but it is
limited to one line and probably due to floating point error for scroll
bar number. It is much less noticeable than the previous jumping which
could involve several pages.
---------
Co-authored-by: WerWolv <werwolv98@gmail.com>
The top margin changes when popups block some portion on the screen.
Previously it was implemented adding blank lines at the bottom which
created problems if the windows was scrolled all the way to the bottom
and the popups were issued. The newer implementation doesn't use spaces
and simply resets the top of the window to be higher making the popup
interaction with the window more natural (no text disappears on the
lines the popup overlaps the text). changes can be identified form the
variables mTopMargin,mOldTopMargin and mNewTopMargin.
Unfortunately those changes revealed problems in cursor navigation that
needed to be addressed. Namely, the window will scroll up and down on
the line before last/second line at the top and similar problems for the
left and right. Those changes correspond to the function
EnsureCUrsorVisible()
Fixing those reveled yet more problems with scrolling past the end of
the file using the keyboard which required adding some more variables
and functions to support the correct handling, You find those changes in
the function MoveDown.
I also renamed some variables that had the wrong casing and fixed the
pasting but that is missing one line.
The reason for the error were the text preprocessor added to deal with
tabs was created to process entire files, not just lines or words. In
that context it cleaned trailing new lines if any existed but in the new
role that's not only unwanted but erroneous.
### Problem description
After that was fixed the console started to add two empty lines between
each output line. When splitting a string using new lines you need to
not add a new line to the last line created. Even if the text ends in an
end line, the split screen code is set to not discard empty lines.
### Implementation description
The fixes are straightforward.
I have implemented a fix that makes the scrollbars behave like they do
in VScode. Vertically you can scroll past the end of the file until only
the last line can be seen at the top. The horizontal bar behaves as if
every line was the same length which is the length of the longest line
in the file. Not only this creates a better user experience, but it also
fixes the annoying flicker that occurs when scrolling through large
files. Also, I have switched all the old draw calls to render text to
regular TextUnformatted calls which adds extra stability to the
resulting display.
To implement the behavior I added a dummy widget with the desired
dimensions.
This time it may be for good.
### Implementation description
As requested I added the preprocessing functionality into the pattern
editor. I had to duplicate a few functions and update the c++ library to
version 20. but now I can make sure the pattern editor doesn't see a tab
again. I also removed the preprocessor from where it was before because
it is not needed anymore. The changes were tested using a file that used
tabs for all its white space. The file was pasted into the pattern
editor and imported as well. I both cases no crashes occurred and the
files had no tabs on them.
### Problem description
This PR implements the feature request described in #1995, that
describes a problem with the `Paste` vs `Paste all` commands. Users
could be thrown off by having `Ctrl+V` act as a simple "Paste over
selection", whereas it's generally accepted as a "Paste all".
### Implementation description
<!-- Explain what you did to correct the problem -->
This PR introduces a new setting, called "Paste behaviour" (under the
"Hex Editor" category).
This setting has three values:
- `Paste over selection`: the current implementation for ImHex. Pastes
only over the selection region, in this case pasting only one byte;
- `Paste everything`: allows ImHex's `Paste` to behave like a `Paste
all` when selecting one-byte regions;
- `Ask me next time`: prompts the user for a choice of behaviour
(default value).
*Note: as users generally use `Paste all` when selecting one-byte
regions, calling `Paste` when selecting over two or more bytes is not
affected by this change, and will still behave like the usual `Paste`
command.*
When selecting a one-byte region, and calling the Paste command, users
that have not defined a preferred behaviour in the settings will be
prompted to choose one, using a brand new popup. The popup also allows
the user to cancel, which will not change the settings' value, and will
cancel the paste action altogether.
### Screenshots
The new popup:
The new setting:
### Additional things
I'm not very good with long descriptions, so I'm open to any suggestions
regarding the text that is included in the popup!
I do think however that we should keep a hint indicating that `Paste
all` is always an option, which could solve the issue altogether for
very new users.
---------
Signed-off-by: BioTheWolff <47079795+BioTheWolff@users.noreply.github.com>
Co-authored-by: Nik <werwolv98@gmail.com>
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
### Implementation description
<!-- Explain what you did to correct the problem -->
### Screenshots
<!-- If your change is visual, take a screenshot showing it. Ideally,
make before/after sceenshots -->
### Additional things
<!-- Anything else you would like to say -->
### Problem description
As described in issue #1841, there was a duplicate on the `Chinese
(Simplified)` translation.
After closer inspection, this came from an i18n file (Script loader,
specifically) having a bad ISO code (`zh_CN` instead of the expected
`zh-CN`).
This also caused a big impact on end users, as loading the latter
(`zh_CN`) would cause the whole UI to be displayed in English (the
fallback language), and only the script loader being translated into
Chinese.
### Implementation description
Fixed the i18n code in the Script loader file.
### Additional things
Closes issue #1841
Signed-off-by: BioTheWolff <47079795+BioTheWolff@users.noreply.github.com>
### Problem description
Long error messages were forcing the width of the window to span the
entire screen.
### Implementation description
The fix was sending the long message to the log and outputting a short
message to the 3d visualizer window.
---------
Co-authored-by: WerWolv <werwolv98@gmail.com>
### Problem description
As suggested in #1904, ImHex could benefit from having the capability of
pasting text directly into the editor.
This also complements the "Copy as... ASCII string" capability already
implemented in the software.
### Implementation description
A new shortcut called `Paste all as string` (to resemble the naming of
the `ASCII string` copy option) now allows to paste plaintext directly.
This shortcut is mapped to the `CTRL + ALT + SHIFT + V` keybind.
Internally, a new flag called `asPlainText` has been added to the
`pasteBytes` function, to minimise code changes and streamline code
readability.
The buffer is a simple type cast of the clipboard, without any
modification applied, which is then handed to the provider's `write`
function.
### Screenshots
The new shortcut is visible in the menu, just below the other paste
options:
---------
Signed-off-by: BioTheWolff <47079795+BioTheWolff@users.noreply.github.com>
Fix crash on UNDO/REDO shortcut press when in "title screen"
### Problem description
ImHex crashes when (by default CTRL + Z/Y) undo/redo is pressed when on
"title"/"starting" screen (no file open, tested on Windows release build
and Linux [WSL] from-source build).
This is due to the shortcut's callback being called even if the
`provider` is `nullptr`. (see `createEditMenu` function).
Theoretically, this is prevented by the `enabledCallback` function
passsed to `addMenuItem`. In this case, though,
`addMenuItem` correctly propagates `enabledCallback` to menu item
creation but does not pass `enabledCallback` to
shortcut creation. Thus, when handling shortcuts, `enabledCallback` is
not used at all and the shortcut's callback
can be called in contradiction with its preconditions. (specified by
`enabledCallback`)
### Implementation description
The implementation wraps the callback in a check that decides whether
the shortcut is enabled or not.
(see changed files)
```c++
auto callbackIfEnabled = [enabledCallback, function]{ if (enabledCallback()) { function(); } };
```
This function is then passed along instead of the `function` (shortcut's
callback).
Alternatively, we can check for `nullptr` in the callback directly. This
would require modification of `createEditMenu`'s contents.
(I did not choose this implementation because I do not think it
addresses the root of the issue).
### Screenshots
None
### Additional things
I'm not sure how big of a deal it is but I am unsure whether I can
capture (`[enabledCallback, function]`) by reference or not.
Since the popup is fairly small I opted for a straight addition parallel
to the find/replace. To make code more clear the functions that create
each popup were coalesced and made their interface simpler. That forced
a reorganization of the data processing which translates to a larger
number of changes than usual. Most of those changes are just moving some
action from one function to another.
The old method to identify popups using the size and position of the
window was dropped in favor of one based on child windows and using
their names for a much easier and robust identification.
Added specialized functions to text editor to jump to a line or to given
coordinates with a simple interface that simplifies older code that
performed the same task.
Because this PR modifies heavily the same code as the previous PR (1983)
it is also included here to make merging easier.
---------
Co-authored-by: Nik <werwolv98@gmail.com>
Errors printed in the console can be clicked to have the cursor jump to
the source code line where the error is at.
The mouse cursor changes its shape to indicate which parts of the error
message can be clicked on the console. When the cursor jumps, the text
editor takes the focus away from the console and it scrolls the window
to make the line with the error is visible if it isn't. This code uses
the function created for the go-to PR but adds code to switch focus to
the target. When the codes are merged please keep both the part that
jumps the cursor and the part that sets the focus.
---------
Co-authored-by: Nik <werwolv98@gmail.com>
### Problem description
As described in #1846:
- the `Edit the Hex` achievement doesn't unlock when it should
- the `ROM Hacks` achievement is not using event-driven architecture
(the functions call `unlockAchievement` themselves)
### Implementation description
Firstly, for the `Edit the Hex` achievement:
- replaced the old event listener on `EventPatchCreated` with a listener
on `EventProviderDataModified`, which picks up bytes changes
- ensured the provider data change comes from a File provider, else
unlocking the achievement wouldn't make sense
- *Note*: a discovered side effect is that the "Fill" function modifies
the provider byte per byte (with a for loop)
- there is no use in testing the size of the data change, as it is
always 1 byte
- the Fill function could probably be reworked to fill in whole regions
at a time?
About the `ROM Hacks` achievement:
- implemented the new, still unused `EventPatchCreated` event.
- signal signature is `const unsigned char *, u64, const IPSKind`:
buffer pointer, buffer size, and IPS kind (IPS/IPS32)
- make use of the `::post` and `::subscribe` methods on said event to
unlock the achievement
- **WARNING::behaviour change**: the event's `post` signal has been
moved in the success branch of the IPS generation condition, meaning
that achievement will only unlock if IPS patch export has worked. I felt
it would make more sense than unlocking an achievement on an error, if
there was any to raise.
---------
Signed-off-by: BioTheWolff <47079795+BioTheWolff@users.noreply.github.com>
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
This PR implements the feature request #1781, that suggests adding a
button to export disassembled instructions into an ASM file.
### Implementation description
This adds a button to export the current disassembled instructions to an
ASM file. Said file is suffixed by an `.asm` extension if not specified
at file creation.
*Note: the file is written to for every `Disassembly` item in the
vector, as it was the easiest and most memory-conservative way of doing
it.*
The file creation task is implemented based on IPS patch exports, so it
fits the same pattern.
A `ToastError` is raised when the ASM export could not complete
successfully.
Translations have been implemented for both `en_US` and `de_DE` for the
two new keys:
- `hex.disassembler.view.disassembler.export`: file export button
- `hex.disassembler.view.disassembler.export.popup.error`: error popup
text
### Screenshots
The button is disabled when the disassembler is working, or when the
disassembly vector is empty.
Here is a complete breakdown of the visual changes:
### Additional things
As expected, the exporter writes every item's `mnemonic` and `operators`
to the file, producing an output like this:
`example.asm`
```asm
.byte 0x7f, 0x45, 0x4c, 0x46
andeq r0, r1, r2, lsl #2
andeq r0, r0, r0
andeq r0, r0, r0
eorseq r0, lr, r3
andeq r0, r0, r1
andeq r1, r0, r0, asr #32
andeq r0, r0, r0
andeq r0, r0, r0, asr #32
```
---------
Signed-off-by: BioTheWolff <47079795+BioTheWolff@users.noreply.github.com>
Fix the AES ECB mode in the data processor along with some other misc
fixes:
- Fixed nullpointer node not working
- Fixed crypto module incorrectly using mbedtls api
- Fixed crypto module ignoring mbedtls errors
- Fixed silently ignoring of errors in AES node
Some context menu entries that were available as shortcuts were greyed
out. This PR aims to fix them and improve how context menus work for the
text editor and the console. The improvements include:
- automatic focus on right click
- automatic selection on right click. If selected text is right-clicked
then copy, cut and find will use the selection, if no selection is
clicked but there is text were right-clicked, then the word will be
selected and used. If right-clicking empty space copy and cut will be
greyed out and find will start empty.
- similar functionality now exists for the console as well except the
menu has fewer options due to it being read-only.
- added esc to close console context menu
### Problem description
At the moment, attempting to edit files using HexII data visualization
is impossible as the editing fields only allow numeric characters but is
interpreted as a hex value
### Implementation description
I yoinked the params used for the hex visualization edit field.
### Screenshots
### Additional things
Ideally this should instead accept the format used by HexII, but I have
what scientists describe as a skill issue and this at least makes it
possible to use HexII without switching back to hex for editing.
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
This PR addresses issue #491 regarding the inability to open ImHex
project files (`.hexproj`) directly from various UNIX Desktop
Environments.
_NB: This PR's scope is limited to UNIX systems, as I have no access to
MacOS systems and could not find a way to automatically create file
associations on Windows._
### Implementation description
<!-- Explain what you did to correct the problem -->
In order for project files to be recognised as such, we need to create a
new MIME type. According to [RFC 6838 section
3.2](https://www.rfc-editor.org/rfc/rfc6838#section-3.2), additional
application MIME types should be defined with a vendor prefix.
_NB: This is preferred over the deprecated `x-` prefix, although RFC
6838 states that the vendor should be "a well known producer"_
This PR proposes a new mime type for ImHex project files:
`application/vnd.imhex.proj`, and associates it with the file format
`*.hexproj`.
It also implements small changes in the Arch Linux and RPM builds, in
order to correctly ship the MIME XML file during installation.
### Additional things
<!-- Anything else you would like to say -->
The implemented change has been manually tested in the following UNIX
systems and graphical environments:
- Arch Linux, with Gnome 47.2
- Arch Linux, with XFCE4
- Fedora 41 Workstation Edition, with Gnome 47.0
- Fedora 41 Plasma, with KDE Plasma 24.08.2
- Ubuntu 22.04.5, with Gnome 42.9
Which tests the functionality of the following build types:
- Arch Linux's `.zst` packages
- Debian `.deb` packages
- RPM packages
For Arch Linux and RPM packages, this has been tested to ensure that
post-installation hooks (such as regeneration of the MIME database) is
correctly executed (which it is). No further changes than those
implemented in the PR are needed.
---------
Signed-off-by: BioTheWolff <47079795+BioTheWolff@users.noreply.github.com>
Fixes
[https://github.com/WerWolv/ImHex/security/code-scanning/223](https://github.com/WerWolv/ImHex/security/code-scanning/223)
To fix the problem, we need to ensure that the multiplication is
performed using a larger integer type to avoid overflow. This can be
achieved by casting one of the operands to `u64` before performing the
multiplication. This way, the multiplication will be done in the larger
type, preventing overflow.
We will modify the line `index += frameCount *
device->playback.channels;` to cast `frameCount` to `u64` before the
multiplication.
_Suggested fixes powered by Copilot Autofix. Review carefully before
merging._
Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
Even if focus is given to the pattern editor the cursor will not show
until the editor contains printable chars.
### Problem description
I was using the wrong condition to enter the loop that skipped the
default empty document.
### Implementation description
The fix was to check if there are lines of text that need to be
processed even if they are empty.
Even tough the sound visualizer has `channels` as one of its parameters
it wasn't using it properly.
### Problem description
The biggest problem is that at each frame the index was being advanced
per channel frame_count increments. The number of channels also
determines how many graph will be needed to display the graphs of the
visualized sound files. Besides these two problems there were many
others like incorrect playback time, cracking audio, etc. which will not
be mentioned.
### Implementation description
To sample the signal a channel sampler was created based on the one used
previously that returns as many sampled signals as there are channels.
This PR aims hopefully at fixing all the problems encountered, and it
has been tested extensively using `Audacity` exported samples to ensure
the visualizer fidelity on playback and graph appearance.
### Screenshots
### Problem description
This PR aims at making inline visualizers work on the data inspector so
that more rgb encodings can be added as custom pattern language
inspector rows. This was never setup to work because the inline
visualizer rendering function was a private member function.
### Implementation description
In order to be accessible from the inspector class the rendering
function was made public.
### Additional things
Missing still is the tooltip to make it behave like other color entries.
### Problem description
Previous implementation ignored everything after the first newline.
Other changes include:
- Added isEmpty() function that checks if editor has no content.
- replaced asserts with default behavior to avoid unneeded crashes.
- fixed off by one error in DeleteRange()
- some crashes occurred because readily available corrective actions
were not being taken. For example start/end being -1 in DeleteRange().
- At the heart of search/replace is the ability to translate from
indices into the text string to line/column coordinates used for
everything. To this end a function (StringIndexToCoordinates) was added
to do the translation taking utf-8 chars into account. This made the
recently added Utf8BytesToChars function unneeded, so it was removed.
- Removed commented out code that is not useful anymore. Also removed
tooltip code which is also unused.
- Removed unused parameter wrapAround to FindNext().
WARNING: this PR won't compile unless [PR 132 from pattern language
repository](https://github.com/WerWolv/PatternLanguage/pull/132) is
merged first. Some changes here are shared by at least another PR to
this repository but there should not be any conflicts as the shared
changes are identical.
### Problem description
fix: Editing patterns with breakpoints sets behaves unexpectedly. As a
simple example, set a breakpoint and insert a blank line somewhere
before the breakpoint location. The breakpoint will appear to move but
in reality it hasn't. To see this set another breakpoint elsewhere in
the file and the old one will be displayed where it is really located
at.
The reason for this and many other problems with breakpoints is that
currently ImHex keeps two set of breakpoints in text editor and in
evaluator that are independent of each other, ie, changes to one don't
affect the other. This PR aims at synchronizing the two sets through the
per provider breakpoints that exist in view pattern editor.
### Implementation description
It accomplishes this by making the text editor version of breakpoints
the principal source of vectors and the ones in evaluator the effective
version. The first allows one to modify the text around and at the
breakpoint and notify others that the changes have induced changes in
the breakpoint locations. The effective breakpoints allow the insertion
and deletion of breakpoints.
View pattern editor is where breakpoints are updated. It receives
notifications from text editor about changes and then makes sure the
version in evaluator is updated with those changes. View pattern editor
also manages breakpoint addition and deletion so before making changes
it gets a copy of the current ones from text editor, sets the ones in
evaluator, uses the evaluator functions to add or delete breakpoints and
finally sets the text editor version with the new version.
### Problem description
Loading a file on the command line or using the context menu in windows
was triggering an achievement before the achievements file was loaded.
This resulted on the achievements.json file to be overwritten with all
the achievements reset except for the one to open files.
### Implementation description
This PR fixes the problem by introducing a boolean that is used to check
if the file has been loaded.
---------
Co-authored-by: Nik <werwolv98@gmail.com>
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
seem to fix#1809. I couldn't build the app, so hoping that's the right
place to update resulting `Info.plist`
### Implementation description
<!-- Explain what you did to correct the problem -->
### Screenshots
<!-- If your change is visual, take a screenshot showing it. Ideally,
make before/after sceenshots -->
### Additional things
<!-- Anything else you would like to say -->
This code fixes issue #1950 This PR depends on PR [#27
](https://github.com/WerWolv/libwolv/pull/27) on libwolv repo
### Problem description
See description of the problem on the issue linked above
### Implementation description
The code in the PR uses function added to the libwolv repository to
remove tabs if they exist on the provider. Additionally, any pasted tab
on the pattern editor will be translated to spaces automatically. This
code was tested successfully using the pattern posted in the issue.
std::forward ensures that the arguments are perfectly forwarded to
loadScript
By reserving space in the features vector based on the size of scripts,
we can avoid multiple memory allocations during the loop. If an
exception occurs, returning an empty vector immediately clarifies that
no scripts were loaded. Without reservation, each call to emplace_back
could potentially trigger a reallocation if the current capacity is
exceeded, which is costly in terms of performance. This leads to more
efficient memory management and can significantly speed up the execution
time when dealing with a large number of scripts.
### Problem description
fix: pressing F5 (start pattern execution ) while being in a break point
hangs ImHex.
The reason for hanging ImHex was that the shortcut procedure was not
checking if the pattern was already running, and it attempted to start
another one. This makes ImHex wait indefinitely for a lock to be
released
### Implementation description
To fix the hanging of ImHex we check the runtime for current evaluation
and if detected we stop it. In all cases the evaluation is started
again.
### Problem description
Fixes provided for the following unreported bugs.
- Environment variables are set to be per provider but used as if they
are not. When a project is loaded all the environment variables for each
provider are assigned to the first provider making it impossible to add
new ones to the other providers.
- When switching providers, the text editor selection, the text editor
breakpoints, the console text, the console selection and the console
cursor position of the old provider are being assigned to the new
provider
### Implementation description
This PR aims at fixing both errors by:
- using variable defined to be per provider so that they affect their
provider only when necessary.
- creating new per provider variables and using them so that each
provider has their own console, selections and breakpoints.
In order to support the newly added per provided features new functions
were added to the text processor for selections and breakpoints. All the
new per provider variables are defined and used in view pattern editor.
### Problem description
Moving error markers to be underwaved created s couple of errors:
- Markers that were 1 char long and located at the end of the line
didn't show on text editor.
- Markers that had zero length or no stack trace didn't show on text
editor.
### Implementation description
- The first error was caused by an off by one error in the column index.
- There was no implementation for errors that had no location or zero
length. Now errors with no location will be shown at the beginning of
the line where they occur. Errors with zero length will be marked from
their location to the end of the line, therefore, errors with no stack
trace will be marked from the beginning to the end of the line.
### Problem description
The only time textures would update was through the file picker. Once UV
coordinates were used, disabling them didn't have any effect because the
vertex array didn't invalidate the corresponding buffer array. When
`shouldUpdate` is true, the light source needs to know as well,
otherwise it will set the location at 0,0.
### Implementation description
Textures were fixed by creating a member variable that holds the file
name of the texture used in the last model rendering.
Instead of invalidating buffer arrays it is much simpler to define a
default UV coordinate set for the case when no UV is specified. if a
texture is not present then the values of UV will not be used. If there
is a texture it must be a minimum of 1 pixel in size. So we choose the
UV coordinates so that every vertex gets assigned the color at the 0,0
coordinate of the texture.
When `shouldUpdate` is on, we also turn `shouldUpdateLightSource` on.
Also included are some formatting changes that are purely aesthetic.
---------
Co-authored-by: Nik <werwolv98@gmail.com>
### Problem description
The process memory provider currently doesn't function correctly on
Linux due to incorrect handling of the special procfs file
`/proc/<pid>/maps`. I don't know if some of this behavior could vary by
distro and/or kernel version, but I've observed the following issues in
my Ubuntu 24.04 environment.
- The current code in master calls `file.readString()` which attempts to
determine the size of the file by [seeking to the
end](https://github.com/WerWolv/libwolv/blob/master/libs/io/source/io/file_unix.cpp#L148).
However, procfs files don't have a defined size, so this fails with a
return of -1. libwolv [interprets this as the file size and attempts to
allocate an enormous
buffer](https://github.com/WerWolv/libwolv/blob/master/libs/io/source/io/file.cpp#L30),
which results in an exception, so ultimately the process memory provider
is unusable on the current code.
- The previous version of the code that went out in 1.35.4 was calling
`readString` with a fixed maximum size of `0xF'FFFF`. This avoids the
seek issue, but when working with special files, a single `read` call
isn't guaranteed to read the requested number of bytes even if that many
bytes are available. In practice, on my machine, this call only ever
reads the first few dozen lines of the file. So the feature works in
this version, but it's unable to see the vast majority of the process'
address space.
- On a more minor note, on rows in the `maps` file that have a filename,
the filenames are visually aligned by padding spaces between the inode
column and filename column. ImHex includes these spaces as part of the
filename, resulting in most of the path being pushed out of the visible
area of the window.
### Implementation description
- To ensure the entire `maps` file is read, I've changed the code to
read from the file in a loop until we stop getting data. I've also set a
fixed limit on the maximum number of bytes to read in one go to avoid
issues with trying to determine the file size.
- I've added a `trim` call to remove any padding around the filename.
### Screenshots
Exception in `file.readString()` in current code (for some reason this
also causes the window to become transparent):
Abridged memory region list in 1.35.4:
Complete memory region list after this PR:
### Additional things
I was focused on fixing this ImHex feature here, but I wonder if some of
this should be addressed in libwolv. Maybe `readBuffer` in file_unix.cpp
should read in a loop until it has the requested number of bytes or
encounters EOF/error?
---------
Co-authored-by: Justus Garbe <55301990+jumanji144@users.noreply.github.com>
### Problem description
The bug can be reproduced as follows:
1) Using Ctrl-F do a search for a term that occurs several times in the
file. and press F3 to visit them.
2) Go back to the fop of the file and insert 3 blank lines.
3) Pressing F3 once or many times will not find the term entered above.
### Implementation description
The reason for this bug is that the positions of the matches are not
being reset when changes can potentially move them.
The fix consists on resetting the search locations when changing the
contents of the file and redoing the search after the changes are made.
The bug was specially problematic when doing replace because the
replacement position would be identified as a match. This PR fixes
replace as well.
---------
Co-authored-by: Nik <werwolv98@gmail.com>
### Problem description
Currently, the pattern editor does not remember where the cursor is
located in each provider. For example, suppose you have 2 providers in
your project, and you scrolled down to line 200 in the first pattern to
make some changes and remembered that the code you want to insert is in
the second provider. Then you switch to the second provider, look for
the code and find it in line 235. Switch back to the first one, and you
are at the beginning of the file. So you again look for the line to edit
paste it to realize that it needs code a few lines before the place you
found it. You switch to the second provider, and you are at the top
again. This gets annoying very fast.
### Implementation description
This PR ensures that, when you return to the pattern in the editor for
any of the opened providers, the cursor will still be at the same place
it was when you switched to a different one. Each provider pattern saves
its cursor position and returns to it when you switch to that provider.
It does that by creating a PerProvider variable and using it when
providers are first opened to set it to the origin and when switching
providers it first saves the position of the old provider and then loads
and sets the saved position of the new provider.
---------
Co-authored-by: Nik <werwolv98@gmail.com>
changes needed were:
1) add an extra dummy empty widget so that width of child and parent are
not always the same. 2) Width of reset button needs to be adjusted when
the width of the child is decreased so that the parent is decreased too.
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
### Implementation description
<!-- Explain what you did to correct the problem -->
### Screenshots
<!-- If your change is visual, take a screenshot showing it. Ideally,
make before/after sceenshots -->
### Additional things
<!-- Anything else you would like to say -->
This is a simple fix for a simple issue. Just check if the last char in
the selection is a double quote and if it is, make the selection one
char shorter.
---------
Co-authored-by: Nik <werwolv98@gmail.com>
### Problem description
The bug described in #1663 was caused by an optimization of the display
rendering that uses drawlists to detect changes. The changes in the 3-d
visualizer don't contain drawlists when axes are turned off.
### Implementation description
The fix is to identify the 3d visualizer among the command lists of the
main window and, if found, avoid skipping frames regardless of the
result of the comparison of drawlists.
Closes#1663
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
I was unable to use the LZMA de-compressor even though 7zip was able to
identify them as LZMA compressed and decompress them without a problem.
Under the debugger the `lzma_code()` call was returning
`LZMA_MEMLIMIT_ERROR`.
### Implementation description
I found online that the error can be fixed using `lzma_memlimit_set()`
using a value obtained with `lzma_memusage()`. I tried to avoid having
to call `lzma_code()` twice but if the functions are called before the
first `lzma_code()`the values of memory obtained fall short and error
occurs again.
I suspect that there are better ways to deal with this other than the
code proposed in this PR, but I haven't been able to find any.
---------
Co-authored-by: Nik <werwolv98@gmail.com>
### Problem description
There are some obvious constrains in the various array data sizes that
we can use to determine if their sizes and in some cases their values
are correct.
### Implementation description
To test their validity, the most important of the various arrays are the
indices, so their definition and use has been simplified a great deal.
In order to treat all variables in a similar manner the changes to the
uv variable posted in an earlier PR were also added here.
With this one, all the recently opened PR's combined should contain
exactly the same changes and fixes as the older 3-d visualizer PR #1850
which I will close shortly after this one is opened.
---------
Co-authored-by: Nik <werwolv98@gmail.com>
### Problem description
Currently when errors are found the entire line where the error occurred
is highlighted and one has to look at the error message in order to find
where the error is located on the line. With this PR the line will no
longer be highlighted and the location of the error will be marked with
an red waved line under the error location. Hovering over the text where
the error occurred produces an error overlay so if several errors occur
on the same line they can all be seen separately.
### Implementation description
The definition of error marker was switched to include column and size
as well as line and message like before.
This change required changing the way view pattern editor draws the
error markers because the errors themselves don't have size information.
Also, a new errorHoverBoxes type was defined to help in the detection of
the floating error messages when error is hovered.
Note that the underwave code depends on having a monospaced. If font is
not monospaced the underwaved text can be short/long or displaced.
### Screenshots
### Problem description
There are some recent issues about Mac keys not configured properly for
the pattern editor. This PR moves all the shortcuts to the shortcut
manager, so they can be edited at will. Even if the key is not
identified correctly it should be possible to use preferred keys for any
action.
---------
Co-authored-by: Nik <werwolv98@gmail.com>
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
https://github.com/WerWolv/ImHex/issues/1895
### Implementation description
Added code that handles streamed zstd data. It is based around the
[official
documentation](http://facebook.github.io/zstd/zstd_manual.html) and the
[example](https://github.com/facebook/zstd/blob/dev/examples/simple_decompression.c)
provided at the main zstd repo.
The loop around the non-streamed version was also removed because I
don't think it was doing anything (no `continue`s, `sourceSize` was
always being set to 0).
### Additional things
To test, I generated streamed zstd data with this python script:
```py
import io; import pyzstd;
with open("data.zstd", "wb") as f:
pyzstd.compress_stream(io.BytesIO(b'ab' * 100), f)
```
And then I ran this pattern script:
```
import std.mem;
import hex.dec;
u8 data[while(!std::mem::eof())] @ 0x00;
std::mem::Section data_sec = std::mem::create_section("data_sec");
hex::dec::zstd_decompress(data, data_sec);
```
Inspecting the section shows the correct data:
Co-authored-by: Nik <werwolv98@gmail.com>
### Problem description
Older image format use to store color values in a small lookup table so
that the image could simply store an index to the table. The pattern
language is not designed to handle this sort of operation which makes
this extension necessary to be able to visualize images of this type.
### Implementation description
The changes add an optional parameter to the bitmap visualizer which
holds the color lookup table. If the image contains values that are
outside the range of possible colors then the first color in the map is
chosen. The dimensions of the image can be equal to or smaller than
rows*columns depending on how the indices to the color map are stored.
For example, you can use 4 bit indices which would make the image half
the size (in bytes) but that limits the number of colors to 16. To store
colors in sizes larger than one byte use an array of the appropriate
sized integers.
### Screenshots
other separate, but closely related fixes are:
- fix: The previous fix also solved the (unreported) bug of being unable
to select utf-8 works by double-clicking.
- fix: The move to next/previous word (Ctr-arrow) behaved differently
depending on the direction. I made both the move left/right functions
share a much simpler algorithm and rewrote the find start/end of word
functions share the same code structure.
### Problem description
The code was using the byte index of the match into the utf-8 string to
store the match locations, but the code that sets the selection uses the
char index into the utf-8 string instead. Another problem was that the
search uses the byte index to determine if it needs to find more
matches.
### Implementation description
Both problems were solved by introducing two functions to switch from
coordinates in units of bytes to coordinates in units of chars and vice
versa.
Co-authored-by: Nik <werwolv98@gmail.com>
### Problem description
When this feature was implemented the text editor handled all the
keyboard input so the blink reset function could be called in one place
for all the key presses. Now that shortcuts are done in the shortcut
manager we need to reset the blink state inside all the functions that
can affect cursor movement.
### Implementation description
Every function that moves the cursor now calls the blink state reset
function at the very start. There may be more functions that need this
that haven't been merged but git should show merging conflicts so it
should be easy to deal with.
### Problem description
1. Old Chinese text json, the order is chaotic. It doesn't compare well
with English.
2. Missing a batch of new English strings.
### Implementation description
1. Synchronize the key sequence of English files.
2. Added a missing batch of text, and translate to CN.
### Screenshots
None
### Additional things
None
On many distributions like AOSC OS, Alpine Linux, Arch Linux, etc.,
boost should be searched with name "Boost".
This should fix the packaging issue on AOSC OS.
### Problem description
CMake fails to find system boost when `USE_SYSTEM_BOOST` is set to ON.
### Implementation description
Find Boost.
Signed-off-by: xtex <xtexchooser@duck.com>
### Problem description
Typing `--help` causes ImHex to exit without outputting anything.
Diagnostic messages from glib, ASan, other libraries that might have
something important to say, etc. are also suppressed.
### Implementation description
This effectively reverts 7c1e33dde6, which
was partially reverted only on Windows by code that was left commented
out in f114239f51.
Allowing other libraries to print to stderr may make the output ‘ugly’,
but lots of things print to stderr that are important for figuring out
why something is bugged, like ASan and glib.
### Additional things
### Problem description
Attempting to do an MD5 hash of a large region (e.g. 2 MiB, ``u8
data[0x200000]``) crashes with a segfault.
### Implementation description
In ``hex::plugin::hashes::hashProviderRegionWithHashLib()``,
``hashFunction->TransformBytes()`` is called with an offset of 0,
because it iterates over ``data`` and not the entire region.
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
Because of `s_drawTexture` never being set to true, textures rarely or
never drew in the 3D visualizer.
### Implementation description
Set `s_drawTexture` to true when correct (valid texture file/object).
### Problem description
The fonts list on Linux does not show all system fonts, and does not
show font names at all.
### Implementation description
Use Fontconfig to make the list less bad if Fontconfig headers are
available.
### Additional things
I like fonts.
---------
Co-authored-by: Nik <werwolv98@gmail.com>
### Problem description
ImHex asks users whether they wish to opt-in to network connections on
start (excellent!), then ignores that and tries to connect to the
network anyway when the welcome screen loads (less excellent!).
### Implementation description
Also don’t connect to the network on the welcome screen if it is
supposed to not do that.
### Problem description
WerWolv/ImHex#1814
### Implementation description
1. Remove the flag that wrongly tells `ImFontAtlas` that it owns font
data, and remove that parameter from the function signature entirely
since now it is always `false`.
2. Rename `takeAtlas` to `getAtlas` since it no longer transfers
ownership as of b652565b57.
This PR updates the nativefiledialog submodule and uses its new feature
to set the ImHex main window as the parent of the dialog window. This
ensures that the dialog stays on top of the main window. This is
currently supported by NFDe on Windows, macOS, and Linux/X11.
Linux/Wayland behaves as it did previously due to limitations in NFDe.
Note that macOS file dialogs have already been parented properly as NFDe
previously used the key window (the window currently receiving keyboard
events) on macOS. However, it's probably better to do the correct thing
and pass the main window to NFDe even on macOS.
### Problem description
The file dialog go behind the main window if the main window is clicked
while the file dialog is open.
### Implementation description
Update nativefiledialog and pass the `GLFWwindow*` of the main window to
the library function.
### Screenshots
Before:
https://github.com/WerWolv/ImHex/assets/6948096/589c3401-702a-4b0a-99ed-02d3e4d9080e
After:
https://github.com/WerWolv/ImHex/assets/6948096/8fef4900-eedc-48d5-8a4e-7bd81e37e3c0
### Additional things
I have tested this on Windows and Linux/X11, but did not test this on
macOS. It would be ideal if someone can help with this. (But as far as
NFDe is concerned, macOS `NSWindow*` handles have been tested (with
SDL2) and works.)
Co-authored-by: Nik <werwolv98@gmail.com>
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
Accomodate -> Accommodate
### Implementation description
<!-- Explain what you did to correct the problem -->
### Screenshots
<!-- If your change is visual, take a screenshot showing it. Ideally,
make before/after sceenshots -->
### Additional things
<!-- Anything else you would like to say -->
### Problem description
The default result export functionality of the Find tool is limited to
only exporting data in a nonstandard text format. This PR adds support
for exporting the results in CSV, TSV or JSON format. The PR also
removes the old format.
### Implementation description
I added the classes `ExportFormatter`, `ExportFormatterCsv`,
`ExportFormatterTsv` and `ExportFormatterJson`, with similar
implementations to the pattern data exporters.
~~I also moved the `ViewFind::Occurrence` class into
`hex/helpers/types.hh`, so the exporters can access it.~~
### Screenshots
### Additional things
Another small change I made is moving the "{} entries found" line on the
same line as the Search and Reset buttons. I think it looks cleaner this
way, but if anyone disagrees, I can revert it.
---------
Co-authored-by: WerWolv <werwolv98@gmail.com>
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
When entering for the end value the same value as the start it is not
recorgnized as valid region, because end must be strictly greater than
begin. Due to the +1 in the Region constructor this is not correct,
because the end is included in the range.
### Implementation description
<!-- Explain what you did to correct the problem -->
### Screenshots
<!-- If your change is visual, take a screenshot showing it. Ideally,
make before/after sceenshots -->
After:
### Additional things
<!-- Anything else you would like to say -->
### Problem description
At least on Windows (I have tested), it fails to save a layout on the
non-portable version of ImHex (unless we have an administrator
privilege).
The log (after an attempt to save a layout as "sample") will look like:
| Component | Message |
| --------- | ------- |
| `libimhex` | `Failed to save layout 'sample'. No writable path found`
|
But the underlying problem is platform-agnostic. It can be also a
problem on other platforms in other ways.
### Implementation description
The layout manager incorrectly queried whether the empty path
(effectively the current working directory) is writable before saving
the layout (not each "layouts" directories it queried earlier).
This is the snippet of the root cause.
```cxx
std::fs::path layoutPath;
for (const auto &path : hex::fs::getDefaultPaths(fs::ImHexPath::Layouts)) {
if (!hex::fs::isPathWritable(layoutPath))
continue;
layoutPath = path / fileName;
}
```
Look at the argument we are passing to `isPathWritable`. `layoutPath` is
a default (empty) `std::fs::path` object and will not be updated until
the directory describing itself is confirmed to be writable.
That caused a problem on non-portable version of Windows because:
1. The current working directory is usually the one of the executable
(`imhex-gui.exe`) and
2. That directory (`C:\Program Files\ImHex` by default) is usually not
writable unless ImHex is executed with an Administrator privilege.
The argument to `isPathWritable` should be `path` (containing one of the
`layouts` directories) and this PR fixes so that.
### Screenshots
### Additional things
This issue is hard to notice when developing because, to reproduce this
bug, the current working directory MUST NOT BE writable (usually
writable when we develop, even when we are working on the non-portable
Windows builds).
This PR compress the debug info in the ELF files built.
This has no impact on the packages (e.g. .deb files) because they themselves have compression, but once installed in the filesystem, they this compression will be beneficial
The compression is opportunistic, happens automatically when possible
For some reason, the web version doesn't work with this (most compiler tests after this seem to fail ?) so it is disabled there
More information: https://github.com/WerWolv/ImHex/issues/1714#issuecomment-2131373826
### Problem description
This PR implements some rudimentary Xcode support for building and
editing ImHex.
### Implementation description
#### Problem 1: Xcode is a multi-configuration buildsystem
The project is already rather CMake generator independent, thus it did
not need to change much to support Xcode's multi-configuration paradigm:
By default, CMake generates a `.xcodeproj` in which targets build their
artifacts into the specified `<>_OUTPUT_DIRECTORY`, postfixed by the
currently active configuration. To better fit the existing paradigm, I
instead opted ot introduce `IMHEX_MAIN_OUTPUT_DIRECTORY`. This variable
is equal to the previously used `RUNTIME_OUTPUT_DIRECTORY` when using
other generators, and is changed to include a configuration specific
_prefix_ when used with Xcode.
The result is different output directories when using Xcode, and no
changes when using any other generator.
#### Problem 2: ImHex does not support AppleClang
To allow building the codebase with Xcode, I have introduced
`IMHEX_IDE_HELPERS_OVERRIDE_XCODE_COMPILER`. Specifying this option to
`ON` will force CMake to honor the user specified compiler settings,
even when using the Xcode generator.
In practice this can be used together with the new "xcode" CMakePreset
to build the project with mainline clang using `xcodebuild`, or Xcode
itself by generating a buildsystem like so:
```
cmake --preset xcode -DCMAKE_PREFIX_PATH=/opt/homebrew/opt/llvm@17
```
This solution is of course not without flaws. The inner workings are a
particularly ugly hack, and mainline clang does not implement the
necessary extensions to allow Xcode to index the code. Regardless this
option is useful to enable future work in terms of bundling/signing
macOS applications in the "intended" way using Xcode without additional
source modifications.
#### Problem 3: Vanilla CMake + Xcode = Bad developer UX
By default, the CMake generated `.xcodeproj` is a mess. Tons of targets
are scattered about, and source files are not organized beyond grouping
them into a "Source Files" and "Header Files" group.
Even "Header Files" is missing, because the ImHex build system does not
regard private header files of libraries as sources of a target, and
Xcode does not try to guess this information.
The solution is twofold:
* Additional code has been added which organizes the targets into a neat
folder structure
* Additional code was added behind a configuration flag
`IMHEX_IDE_HELPERS_INTRUSIVE_IDE_TWEAKS` which automatically creates
source file trees in Xcode targets, and discovers the non-declared
header files via the folder convention.
### Screenshots
N/A
### Additional things
As a bonus: `IMHEX_OFFLINE_BUILD` assumes that ImHex-Patterns is cloned
into the source tree. I have added an additional fallback that tries to
locate it as a sibling folder of `${CMAKE_SOURCE_DIR}`, as this meshes
better with my filesystem setup.
The setup was tested with `CMake 3.29.2`, `Xcode 15.2`, and `llvm@17`
from homebrew.
### Problem description
#### Problem 1
In borderless mode ImHex disables the standard macOS titlebar rendering
and input processing. As a result double clicking the titlebar does not
trigger the native macOS behavior set in `System Settings -> Desktop &
Dock -> Double-click a window's title bar to [Zoom/Minimize/Do
nothing]`.
#### Problem 2
The ImHex window shows up as blank/transparent when de-minimizing it
from the dock.
#### Problem 3
Widgets experience ghost hover inputs from the past position of the
cursor during live resizing.
### Implementation description
ImGui elements consume input events in the order they are drawn. As a
result by "drawing" an `InvisibleButton` over the content area of the
titlebar we can catch unprocessed clicks in the titlebar area.
Connecting this button's double clicks to the native window is then a
trivial endeavour.
The blank windows was caused by the rendering stack clearing the GL
buffer, but proceeding to draw nothing in it. I have short circuited
this path.
Ghost hover inputs were squelched by consistently moving the ImGui
cursor to `0, 0` during a live resize. The OS will dispatch a cursor
positioning event once the resizing ends, restoring normal behavior.
### Screenshots
N/A
### Additional things
N/A
---------
Co-authored-by: Nik <werwolv98@gmail.com>
### Problem description
ImHex didn't support Hungarian :(
### Implementation description
I translated ImHex to Hungarian :)
### Translation Coverage
| Plugin | Percentage |
|---------------|------------|
| builtin | 99% |
| diffing | 100% |
| disassembler | 100% |
| hashes | 95% |
| script_loader | 100% |
| ui | 100% |
| visualizers | 100% |
| windows | 100% |
| yara_rules | 100% |
### Additional Notes
There are four Hungarian "special" characters that fall outside the
default Unicode ranges loaded by ImHex, resulting in them being replaced
with the "�" character. These letters are Ő (U+0150), ő (U+0151), Ű
(U+0170) and ű (U+0171), all included in the Latin Extended-A Unicode
block.
The easy fix for this is to include the "Unicode Latin Extended-A" range
when loading the font glyphs in
[init_tasks.cpp:189](99abc4e78a/plugins/builtin/source/content/init_tasks.cpp (L189)).
This change would also unlock the full character range of Bosnian,
Croatian, Czech, Estonian, Latvian, Lithuanian, Maltese, Polish,
Romanian, Slovak, Slovene and Turkish. I can add the commit to this PR
if maintainers are okay with it. **EDIT:** Added a commit that loads
Latin Extended-A by default.
Also note that some words are longer than their English counterparts,
resulting in certain UI labels overflowing from their parent containers,
and being cut off. I might change some of the longer labels to be more
compact in the future, but the container size limitations will have to
be addressed eventually (e.g. with horizontal scrollbars).
---------
Co-authored-by: Nik <werwolv98@gmail.com>
### Problem description
https://gitlab.freedesktop.org/mesa/mesa/-/issues/11135
It turns out LLVMpipe only supports 4x multisampling. Checking
GL_MAX_SAMPLES seems like the right thing to do.
### Implementation description
~~Right now, I only check GL_MAX_SAMPLES. Depending on the format, we
might need to check GL_MAX_INTEGER_SAMPLES. I don't know how likely it
is that you might want to use a different format in the future,
glGetInternalformativ might be a safer option to retrieve the max number
of samples we can use.~~
Ended up implementing it with glGetInternalformativ.
### Additional things
I guess I could merge the ```if```s at lines 95, 99 and 103 in
imgui_imhex_extensions.cpp while we're at it.
---------
Co-authored-by: Nik <werwolv98@gmail.com>
### Problem description
By default, the cursor's blinking cycle is independent from user
actions, which feels unconventional while editing pattern code. This
change ensures that the blinking cycle is reset, and the cursor is made
visible every time the editor receives a keyboard event like character
entry, or arrow navigation.
### Implementation description
I moved the hard-coded blink timing numbers to their dedicated static
constant fields since now they are referenced from multiple different
places.
I added the function `void ResetCursorBlinkTime()` to `TextEditor.cpp`
that will reset the blink cycle to `currentMillis() -
sCursorBlinkOnTime`, ensuring that the cursor is visible. This function
is called for every keyboard and mouse click event.
### Screenshots
Before:
https://github.com/WerWolv/ImHex/assets/45818400/668c6802-79a3-450b-80d3-d6abf2ce27be
After:
https://github.com/WerWolv/ImHex/assets/45818400/ee7f60e0-a75f-416d-b86d-8d12b5cdadf2
---------
Co-authored-by: Nik <werwolv98@gmail.com>
### Problem description
This PR adds a shortcut link on the Search (Ctrl+F) popup to the more
advanced "Find" view that is normally accessible at the `View > Find`
menu.
### Implementation description
I added a simple `ImGuiExt::IconHyperlink` link to the `draw()` function
of the `PopupFind` class to display the hyperlink. Clicking the link
will open the Find view, bring it into focus and close the current
popup.
### Screenshots
Before:
After:
### Additional things
- Localization keys were added to all .json files, but only `en_US` is
populated for now.
### Problem description
Merging my previous PRs, #1660 and #1658 has resulted in a conflict
causing an application crash due to a missing `ImGui::BeginDisabled();`
in the `PopupSelect` class. Specifically, none of the code related to
offset validation made it into the final merge. This PR fixes the crash
by reintroducing the deleted lines.
### Additional things
The nightly release build seems to be unaffected by the crash, most
likely because ImGui's `DisabledStack` assertions are only enforced in
Debug mode. The offset validation was still missing before this fix.
### Problem description
In previous versions of ImHex, all tool windows were implemented as
static popups fixed in the upper left position of the hex view. This PR
refactors all tool popups to use floating windows that can be dragged
around by the user, or closed with a dedicated close button on the title
bar. These popup also support a stylable transparency when the user is
not hovering their mouse over the window.
### Implementation description
I rewrote the logic in `ViewHexEditor::drawPopup()` to use a custom
`ImGuiExt::BeginHoveringPopup` function for rendering the popup windows.
This new function is an almost exact replica of the built-in
`ImGui::BeginPopupModal`, except it does also displays the default
window title bar with a close button.
A second custom function, `ImGuiExt::PopupTitleBarButton` was also added
for rendering small icon-based buttons into the title bar of the parent
popup window. This new function was used to implement an optional
"Pinning" feature that individual popup implementations can specify. If
a window is pinned, it won't close automatically when its main action is
executed. For example, the "Select" button on the Select dialog will
close the popup by default, unless the window is pinned.
### Screenshots
Popup dialogs before:
Popup dialogs after:
https://github.com/WerWolv/ImHex/assets/45818400/99d1a628-8ac1-40ac-9146-9062091bb0db
### Additional things
- When the user stops hovering their mouse over a popup window, it
becomes semi-transparent, making it easier to see the content behind it
- This PR also introduces the `styles.imhex.popup-alpha` style, making
the transparency effect configurable, including the ability to disable
the effect completely by setting `popup-alpha` to `1.0`.
- Fixed a bug that caused some popup windows to ignore the Enter and the
KeypadEnter keys. With this PR, all tool windows will execute their main
action when the user presses either one of the two Enter keys, and will
also close automatically unless the window is pinned.
### Possible changes and improvements
- Should the transparency effect be disabled if a window is pinned?
- Should the transparency factor be modifiable on the Settings/Interface
page?
- A keyboard shortcut could be added for quickly pinning / unpinning the
current window.
- Can the pin icon stay on the left, or should it be moved next to the
close button, with a similar circular background?
---------
Co-authored-by: WerWolv <werwolv98@gmail.com>
### Problem description
This PR offers two improvements:
1) When selecting / jumping to an offset that falls within the current
viewport of the scroll view, the scroll offset will no longer force the
selected byte to the top of the view. Instead, the scroll offset will
only be changed if the selected byte is outside the current view.
2) In case a wrong offset is entered into the Select or Goto dialog
(e.g. and offset beyond EoF), the dialog's button will be disabled.
### Implementation description
For the first change, I modified the logic that recalculates the
`m_scrollPosition ` based on the current byte offset.
For the second change, I added validation logic to both popups to ensure
that the entered offsets are valid (using `provider->getActualSize()`).
In case of the Select popup, I wrapped the button into an
`ImGui::Begin/EndDisabled` to enforce the validation check.
### Problem description
When the close button is clicked, `ImGui::BeginPopupModal()` sets the
bool passed into the second parameter (p_open) to false. However, the
closing logic did not take this into account, making it difficult to
actually close modal popups.
For example, closing the "Export pattern File" modal took several clicks
on the "X" button, now it closes instantly.
### Implementation description
I added an additional check for the `open` variable being `false` in the
logic that checks the closing condition.
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
This PR aims to address #1645 that caused the built in file provider's
change monitor to trigger the notification popup dialog multiple times
in a row after multiple external file changes.
### Implementation description
I added an additional boolean field
`m_changeEventAcknowledgementPending` that tracks whether there are any
pending or unacknowledged change notification dialogs to prevent further
dialogs from being opened. The flag is only reset to its initial value
once the user has acknowledged the first `PopupQuestion` dialog.
Since the file is reloaded only after the user clicks 'Yes', it is
unnecessary to ensure that only the latest popup is acknowledged.
- Better argument parsing
- Allow processing all language folders at the same time
- Allow an optional reference language when translating
- Save translations on KeyboardInterrupt
- Fixes a ooold input issues by importing readline
(https://github.com/kovidgoyal/kitty/issues/6560)
- Add untranslate mode to remove translations by a key regex
---------
Co-authored-by: Nik <werwolv98@gmail.com>
### Problem description
The framerate limiter doesn't work when inputs are being sent (eg mouse
cursor moving over the window), because `glfwWaitEventsTimeout` returns
early when it encounters an event.
### Implementation description
I made it sleep for the remaining time when that happens.
## Add dependency download script for OpenSUSE Tumbleweed.
### Testing
After installing the dependencies, I have managed to successfully
compile the project as described in the building instructions:
https://github.com/WerWolv/ImHex/blob/master/dist/compiling/linux.md
I have tested the script on a fresh installations of the OpenSUSE:
- Tumbleweed Server
- Tumbleweed Generic Desktop
### Note
Dependencies added in the script are based off of scripts for other
Linux distributions.
Because of that, some download dependencies might not be necessary (or
could be already included by other packages), so it is very much
possible that some of them could be removed.
This pull request fixes build on FreeBSD. The changes are conditioned
with `#if defined(__FreeBSD__)` preprocessor macro and they should not
affect build for other operating systems.
---------
Co-authored-by: Nik <werwolv98@gmail.com>
Co-authored-by: iTrooz <hey@itrooz.fr>
Fixed console error messages using doc comment syntax highlights. Fixed
results of find not updating when march case was toggled. Fixed syntax
highlights of nested ifdefs. Fixed editor cursor blinks if OS focus goes
to another window. Fixed Highlights of "\\\"" was incorrectly handled.
---------
Co-authored-by: Nik <werwolv98@gmail.com>
This PR adds a test architecture to be able to test plugins
Main infrastructure done by @WerWolv
---------
Co-authored-by: WerWolv <werwolv98@gmail.com>
### Problem description
WASM build does not support copy/paste beyond the application. Meaning,
there's no practical way of sending text back and forth across the
application border.
There are lengthy threads why this is a technical challenge in
WASM/Browser world, e.g:
- https://github.com/pthom/hello_imgui/issues/3
- https://github.com/emscripten-core/emscripten/pull/19510
### Implementation description
Implements a workaround solution as Header only C++ library, as proposed
and implemented at:
https://github.com/Armchair-Software/emscripten-browser-clipboard
Maybe there are cleaner ways of achieving the functionality. Definitely
would like to have some discussion around this. 👀ℹ️ The proposed PR "works for me" on Windows, using CTRL-C/V shortcuts
to copy text from and to the application. On MacOS the system shortcut
to Paste is different from what ImHex has defined. This results in
system Paste shortcut of command-V triggering the browser callback to
synchronise the application clipboard, but no actual Paste takes place
within ImHex.
If there would be a clean way to trigger the paste command, that would
be wonderful (or get the context and references to write the data to the
cursor, but I was unable to find a clean solution). The only proposed
solutions in the referenced threads were about triggering paste event
internally via Key events. This seemed wonky 🙃 , so is not currently
implemented. At the moment the paste on MacOS is command+V followed by
control+V.
### Additional things
This is definitely a stopgap solution before the ImGui and Emscripten
take a more proper solution in enabling Copy/Paste outside the
application borders. However, I feel like this is a must have capability
to make the WASM build more useful, not just for trying out ImHex.
Cheers! 🍻
---------
Co-authored-by: Nik <werwolv98@gmail.com>
Prior to this, at least on Linux/MacOS/etc, I guess it was not possible
to compile with Zstd included for `hex::dec::zstd_decompress()`:
Every other target lib for the decompression plugin would compile,
**except for** Zstd. Additionally, the target name `zstd` caused CMake
to not be able to find `Find(ZSTD).cmake` due to case-sensitivity
differences between Windows and Unix-based file paths, of course.
With that said, I'm not too sure if this will break building w/ Zstd on
Windows... Hopefully not 😄
d511080814 introduced a regression where
running imhex with arguments (i.e. `imhex blob1.bin blob2.bin`) will
segfault due to a null pointer dereference.
### Implementation description
This patch updates `getSubCommands` to follow the same control flow as
`getFeatures`, where if the function returns a null pointer, it will
return gracefully rather than crash.
I saw that diffing is not mentioned anywhere in the README.md and people
I referred ImHex to did not consider using the editor due to diffing
being a requirement for them.
### Problem description
Diffing is not mentioned anywhere in the README.md
### Implementation description
Added Diffing to the features list of README.md
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
Implement a Linux backend for the ProcessMemoryProvider plugin.
### Implementation description
<!-- Explain what you did to correct the problem -->
Most of the provider code is the same between Windows and Linux. The
primary differences are:
- enumerate PIDs in `/proc/` to get the process list
- use `/proc/<PID>/cmdline` as the process name
- parse `/proc/<PID>/maps` to get the module list
- reading/writing from memory is done using
`process_vm_readv`/`process_vm_writev`
NOTE: `sudo setcap CAP_SYS_PTRACE=+eip build/imhex` must be run to give
the binary permission to read another process' memory. Running as root
user should also work but I would not recommend it.
### Additional things
The existing translations keys no longer match since I moved the plugin
from `windows` to `builtin`.
I'm not well versed in C++ so I attempted to keep my changes rather
simple. Feedback is very welcome.
---------
Co-authored-by: WerWolv <werwolv98@gmail.com>
### Problem description
Commit 1249eb3261 added `#include
<GL/gl.h>` in a file, but MacOS uses `<OpenGL/gl.h>`.
### Implementation description
If compiling on MacOS, use `<OpenGL/gl.h>`. Otherwise, use `<GL/gl.h>`.
### Additional things
There is a redefine, so I used the preprocessor to ignore it for MacOS.
This feels a little awkward to me (but I don't know how to do it
better).
I'm not particularly familiar with C++, so let me know if this is the
right solution. This is also my first pull request, so feel free to
roast me for anything else :)
I said I'll do it in #1414
Warning: the output zip does not pass SIP (System Integrity Protection),
and I have no plan to do it for now. So if someone here has a Mac, feel
free to to try to fix the problem (which is probably our build system)
Link #1109
---------
Co-authored-by: Nik <werwolv98@gmail.com>
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
Before this PR, the CRC-32 value of ImHex is not like most of others.
### Implementation description
<!-- Explain what you did to correct the problem -->
Just set m_reflectIn and m_reflectOut of CRC-32 true by default.
It turned out that the achievement "ROM hacks" wasn't actually
unlockable.
I'm not sure if the IPS32 patch also has to trigger this achievement,
but it seemed logical to me so I added the call to both methods.
Fix build problem introduced by
https://github.com/WerWolv/ImHex/pull/1398
I didn't refactor the code well because I thought `CREATE_PACKAGE` was
an option only used for Windows, when it needs to be used for MacOS too
Adds a reference to [Pcap
Plugin](https://github.com/Professor-plum/ImHex-Plugin-Pcap) under
Third-Party Plugins
### Problem description
Adds a reference to an external repository for an ImHex Plugin to view
raw data within pcap files.
### Implementation description
The ImHex Plugin is implemented as an additional Data Provider. It
allows BPFs and selecting multiple packets for data reassembly.
### Screenshots
### Additional things
Should there be any questions I'm also available on Discord.
Testing various fonts and sizes I realized there were still problems
with the layout of the bit labels. Also I reorganized the code so that
lambdas are defined just before they are used. Comments needed
punctuation too.
Part of the problem was that I had been assuming all along that the
check boxes were ImGui's originals which are always square. In actuality
the width is determined by the width of the character for '0'. Also
ImGui table was adding spacing to separate cells which made the boxes
not start at same place as column. Also for some reason using indent of
zero didn't work as expected but using 0.1 pixels worked. With those
problems fixed it is fairly easy to make sure the labels are centered at
the box except when the first mantissa checkbox gets a label which is
wider that the box width. Before and after show results for different
fonts.
Before:
After:
### Implementation description
I synchronized the json key order with `en_US.json` in each lang file.
### Additional things
Here are simple python script that make this change
```python
import json
from collections import OrderedDict
original = "./en_US.json"
modified = [
"./de_DE.json",
"./es_ES.json",
"./it_IT.json",
"./ja_JP.json",
"./ko_KR.json",
"./pt_BR.json",
"./zh_CN.json",
"./zh_TW.json",
]
for modify in modified:
dict_2 = {}
with open(original, 'r', encoding='utf-8') as f1, open(modify, 'r', encoding='utf-8') as f2:
dict_1 = json.load(f1)
dict_2 = json.load(f2)
dict_1_translations = dict_1["translations"]
dict_2_translations = dict_2["translations"]
ordered_dict_2 = OrderedDict((k, dict_2_translations[k]) for k in dict_1_translations.keys())
dict_2["translations"] = ordered_dict_2
with open(modify, 'w', encoding='utf-8') as f2:
json.dump(dict_2, f2, ensure_ascii=False, indent=4)
```
This PR integrates my libmagic patch into the Dockerfile rather than my
vcpkg fork (so we can use the official vcpkg repository in the future)
That said, we still use my vcpkg fork for now because official vcpkg
fails to compile freetype
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
Fixed possible bug of `EventManager::unsubscribe`
`std::map` only allows unique key, but the same token can subscribe to
multiple events.
1a2a926b77/lib/libimhex/include/hex/api/event.hpp (L104-L107)
If the previous token has already subscribed to an event, then when
subscribing again, `getTokenStore().insert` will not do anything
(Because its type is `std::map`)
1a2a926b77/lib/libimhex/include/hex/api/event.hpp (L122-L134)
At this point in `unsubscribe`, the `iter` may not be able to find the
correct event and erase it
### Implementation description
<!-- Explain what you did to correct the problem -->
Change `tokenStore` to `std::multimap` instead of `std::map`, which
cannot unsubscribe multiple events correctly
### Screenshots
<!-- If your change is visual, take a screenshot showing it. Ideally,
make before/after sceenshots -->
### Additional things
<!-- Anything else you would like to say -->
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
When starting ImHex on a MacBook model with both integrated and discrete
graphics, it will force the computer to use the discrete graphics card.
This causes increased power usage, meaning the fans will spin up, the
battery will drain faster, etc. This program is not very graphics
intensive, so using the discrete graphics card shouldn't be needed.
### Implementation description
<!-- Explain what you did to correct the problem -->
I changed the
[`GLFW_COCOA_GRAPHICS_SWITCHING`](https://www.glfw.org/docs/latest/window_guide.html#window_hints_osx)
setting in GLFW to not enforce using the discrete graphics.
### Screenshots
<!-- If your change is visual, take a screenshot showing it. Ideally,
make before/after sceenshots -->
### Additional things
<!-- Anything else you would like to say -->
My editor is configured to automatically remove trailing whitespace, so
I hope that those whitespace changes are ok
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
### Implementation description
Updated Traditional Chinese (Taiwan) translations for ImHex.
### Screenshots
<!-- If your change is visual, take a screenshot showing it. Ideally,
make before/after sceenshots -->
### Additional things
<!-- Anything else you would like to say -->
### Problem description
The issue seems to only affect the web version of ImHex but it is not
clear why. The labels of the checkboxes that represent bits in the IEEE
745 floating point tool were lay out incorrectly.
### Implementation description
The new code ensures that the bit labels are centered and located in the
middle of the checkbox regardless of the font size and the dear imgui
sized checkboxes.
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
This might fix building with capstone 4 as discussed in
https://discord.com/channels/789833418631675954/1155669027306340393/1155669027306340393
### Implementation description
<!-- Explain what you did to correct the problem -->
moves the max definition inside the if statement that checks for
capstone 5
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
This will simplify the codebase in the pattern drawer. It returns early
on conditional statements and reduces the amount of nested code making
it easier to read and track while developing.
I noticed the bad score on code factor so I reorganized it to make it
more readable and maintainable. In order to break down the big function
into it much smaller parts I encapsulated all the variables that the
functions need to access in two classes, one for the imgui related
statics and the other for non-static variables.
When writing the smaller functions I was noticed that there was room to
simplify the existing algorithms by writing functions that could be
called by parts that previously shared no code. I tested the changes the
same way I tested the original and it seems to work the same way but
maybe a bit faster. Although it may be possible to further optimize the
present code code factor no longer flags the function at all.
As discussed (many times) on Discord, does the same as the new favorite
tag, but instead allows you to add multiple groups.
Initially, this would cause some insane issues with draw/reset
(apparantly) fighting eachother in the pattern drawer. After a lot of
trial and error, I decided to rewrite the flow that is responsible for
calling reset. Now evaluating patterns is the one to decide when the
reset happens, not the core "game"-loop.
To make sure that draw and reset can never happen at the same time, the
mutex originally used for the favorites has been repurposed. Due to the
restructuring, the mutex in the favorite-task is no longer needed, as
that will only ever kick-off after reset is called and if there are
actually patterns, which can never line up to be accessed on different
threads at the same time.
Last but not least, I noticed that hard crashes could result in your
config file getting overridden. I added a check to prevent that.
Last I issue I can see is that if you use an excessive amount of
favorites/groups, a crash can still happen, but it only happens when you
close the program (occasionally, but unpredictable). Before, this would
happen if you ran the evaluation a second time. I boiled the cause of
the crash down to these lines of code in evaluator.cpp >
patternDestroyed:
```cpp
if (pattern->isPatternLocal()) {
if (auto it = this->m_patternLocalStorage.find(pattern->getHeapAddress()); it != this->m_patternLocalStorage.end()) {
auto &[key, data] = *it;
data.referenceCount--;
if (data.referenceCount == 0)
this->m_patternLocalStorage.erase(it);
} else if (!this->m_evaluated) {
err::E0001.throwError(fmt::format("Double free of variable named '{}'.", pattern->getVariableName()));
}
}
```
Specifically, trying to access the `*it` is the reason for the crash
(this was also the cause of the crashes before my fixes, but then during
evaluation).
I'm suspecting the root cause is somewhere in the `.clone` methods of
the patterns. I'd say that for now a crash when closing the program is
more acceptable than during evaluation (which can even happen if you use
favorites).
From #1265, Looks like Error Popup doesn't handle properly in some
circumstances.
---------
Co-authored-by: iTrooz <hey@itrooz.fr>
Co-authored-by: WerWolv <werwolv98@gmail.com>
Fixes random .NET plugin crash caused by the console window being freed
on the same process.
### Problem description
Sometimes when launching from the explorer ImHex crashes
### Implementation description
We've implemented a launcher to avoid freeing the std streams, that
generated the bug.
### Screenshots
N/A
### Additional things
Blame windows console subsystem
### Problem description
Ref #1210
### Implementation description
Call
[`SetFileTime()`](https://learn.microsoft.com/en-us/windows/win32/api/fileapi/nf-fileapi-setfiletime)
everytime `FileProvider::save()` is called.
### Additional things
I moved the call to `File::close()` from `FileProvider::open()` to
`FileProvider::close()` because `SetFileTime()` requires a file handler
as input, so I need `File::m_file` to be valid.
As requested on Discord, a PR to revert the FPS limiter back to the one
that was used in v1.30.1.
The new FPS limiter seems to be flawed in that it runs at about half the
speed it is supposed to be.
See this illustration:
Left is v1.30.1, right is the new version (without this fix). See how
long it takes to respectively reach 0xE90.
This is not a performance issue, because when you fully unlock the
framerate on the right, it's just as fluent as on the left.
Currently, LTO isn't enabled (at least on Linux ?), because LTO doesn't
recognize any language as being enabled.
I fixed that by explicitly enabling C and CXX languages before enabling
LTO
This PR fix libmagic dumping files in the imhex cwd when compiling them
This code was actually written by you (notice the source branch), this
PR is just a reminder that the fix works and you can merge it ^^
---------
Co-authored-by: WerWolv <werwolv98@gmail.com>
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
There are unused ImHex-specific flags in the custom GLFW build on MacOS.
Also, it isn't built in release mode
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
<!-- Describe the bug that you fixed/feature request that you
implemented, or link to an existing issue describing it -->
see #723
### Implementation description
<!-- Explain what you did to correct the problem -->
`xdg-desktop-portal-wlr` is not relevant because it does not provide the
FileChooser interface.
`xdg-desktop-portal` needs the `WAYLAND_DISPLAY` env var on wayland and
the `DISPLAY` and `XAUTHORITY` env vars on xorg. If the dbus user
session bus is not run in a way that it already gets those variables,
they must be given via `dbus-update-activation-environment(1)`.
### Screenshots
<!-- If your change is visual, take a screenshot showing it. Ideally,
make before/after sceenshots -->
### Additional things
<!-- Anything else you would like to say -->
fixes#723
<!--
Please provide as much information as possible about what your PR aims
to do.
PRs with no description will most likely be closed until more
information is provided.
If you're planing on changing fundamental behaviour or add big new
features, please open a GitHub Issue first before starting to work on
it.
If it's not something big and you still want to contact us about it,
feel free to do so !
-->
### Problem description
- Fixed disk provider not working for linux
### Implementation description
- Used ioctl instead of fstat
- Fixed buffer issues
---------
Co-authored-by: WerWolv <werwolv98@gmail.com>
This PR intends to add support for .NET scripts that can extend ImHex's
functionality in a portable and cross-platform way.
---------
Co-authored-by: Justus Garbe <55301990+Nowilltolife@users.noreply.github.com>
This PR it just a hack to fix#1160 , it doesn't solve the underlying
problem.
It fixes the problem because by using execvp() directly, it avoids the
call to `sh` done with `system()`, which has a bug on Ubuntu 22.04 which
makes it i,compatibles with the glibc inside the AppImage.
It doesn't fix the underlying problem because the programs we call
themselves still link to the AppImage's libraries instead of the system
ones.
### Problem description
Currently, the providers use the method `isSavable()` to determine both
if they can use "Save" or "Save as".
This behaviour is problematic because some providers may need to be
saveable but not saveable as: for example the view provider. The
original provider may not allow to be saved.
### Implementation description
I separate these two behaviour by creating another function:
`isDumpable()`, that return true by default but can be overridden by the
provider to return false, if the provider should not be dumped in any
way.
### Additional things
While I was at it, I also marked "export" operations as needing the
"dumpable" flag. That way, we can't accidentally export the whole
address space of a process as base64.
I also added documentation for these some functions in Provider
This PR fixes some things about crash handling:
- when the terminate handler is called, immediately set it back to the
original one, so can't make a recursion if the crash-handling code fails
- Only save projects if the crash occured after Imhex finished startup
- do not update the project location when saving the crash backup file:
this will remove problems when `EventAbnormalTermination` is called
before `crashCallback()`
I also added a bit more documentation
In order to do this I add to make some other additions :
- Add a warning popup (TODO, maybe add some icons to differentiate
error/warning popups in a future PR ?)
- create showError() and showWarning() functions, as helpers to show a
message both to the logs and as a popup
This PR updates imgui and its dependencies from the last time, ~1 year
ago
(deabacbd50)
Commits will be refactored before merging
Things you might ask :
- why did you remove `ImGui_ImplGlfw_SetBorderlessWindowMode()` ?
-> Where is it used ? The only usage of it I see is commented
(cb9a3b1f55/lib/external/imgui/source/imgui_impl_glfw.cpp (L757))
- why did you remove the implot anti aliasing flag ?
-> They.. seem to have removed it altogether ?
https://github.com/epezent/implot/issues/479
This PR does two things :
- save opened projects as recent entries
- refactor stuff about recent entries in a separate file. The reason is
that I felt like welcome_screen.cpp was really big ( 685 lines before
this, 500 now). What do you think ?
---------
Co-authored-by: Nik <werwolv98@gmail.com>
Really small issue, in Help->About->Imhex directories, there would be an
extra entry at the end because you seemed to loop over the
fs::ImHexPath::END entry
- Add a new file 'crash.json' to store metadata about the crash, like
the log file or project opened
- show the log file of the session that caused the crash to the user
- Correctly restore the project path
Compiler when compiling doesn't use them in those cases and emit a
warning, which is turned into an error by `-Werror`. Unfortunately,
CPack doesn't expose the logic it uses for stripping binaries.
I just realized one feature request existed about this tool and have
added a comment to it referring this pr. Errors and additions are
described in the fork commit already. I'm not sure if I should repeat
them here again. I have tested the changes thoroughly, but it is always
possible some fringe case was not tested and is incorrect. The tests
were done using the many similar online calculators for IEEE 754
floating point formats.
IEEE 745 floating point tool redesign modeled after 'float toy' web app
(http://evanw.github.io/float-toy/)
Streamlined output using colors and compact layout which can be further
simplified.
Chosen display mode (detailed or simplified) is automatically saved and
set on new sessions.
Edit the binary bits, the integer hexadecimal or the floating point
decimal values and the entire app will update with the change.
Supports the main IEEE745 standard formats (half, single and double
precision) together with custom formats of size <= 64 bits.
Each format choice uses and displays the number of significant decimal
digits defined by the mantissa size.
Added labels to identify the location of each bit box inside the binary
representation.
Satisfies round trip / idempotent (reproducing) conversion property
Added theme colors, radio buttons for display mode and a clear button
that resets the tool.
Removed previously and incorrectly added locale translation to various
labels and languages
Attempted to adhere to code style formatting using existing code as
example.
An effort was made to use preferred variable types and functions from
std namespace when appropriate.
Attempted to document code using comments.
Not implemented / left to complete at an later time
Arbitrary width and precision formats.
Extended precision formats.
Shortest string property.
hexadecimal floating point display and conversions.
This is aimed at use by linux distros, where package updates come from a
central location, and users shouldn't need to worry about updating ImHex
on their own. This disables parts of the ImHex UI that would not be
useful in that case.
Tested and confirmed that this works in both states of the of the
`-DIMHEX_DISABLE_UPDATE_CHECK` switch.
This PR adds some documentation. It's actually pretty random, I followed
the function calls I was curious about and commented whenever I wasn't
sure/I thought it needed clarification
You might want to make sure to squash them, because the commits are kind
of a mess, I didn't went through the effort of interactive rebase
This PR does two correlated things:
- Show a generic error message (Failed to open provider) rather than a
file-specific message (Failed to open file) when a provider fails to
open
- Set the error to something more specific when opening a file fails
Currently there is no way to save the pattern code progamically from a
plugin unless the builtin plugin is modified to add those events. This
pull request will be adding ability to load and save pattern code from
specified file.
This PR handles errors that NFD might encounter (both in Init() and the
other method to open the dialog), and log them in the logs and in the
GUI
This (among other) fix the crash I had running ImHex as root and opening
a file
This pull request changes Config Directories on Linux to only include
the XDG_CONFIG_HOME directory, as opposed to all directories in
XDG_DATA_DIRS before (introduced in
https://github.com/WerWolv/ImHex/pull/644/files#diff-c1a4d2b63fed168a9a3568944e9cadeae096f2ddcec3649e4a9b2d29fd104be0L162-L166).
Reasons:
- This changes the location of the config file to the standard directory
meant for configurations
- This prevents the config file from being read/written in system
locations, like /usr/share
This PR also includes a migration task that will run on Linux and move
config/GUI dimensions to the new directory
as a bonus, as discussed on discord, it writes the logs to a Data
directory instead of a Config directory
PR title is self explaining
I may modify other providers implementations to display a detailed error
message later
I'm not sure how to deal with other locales because the format changed.
Before, I had to add and comment the key in all locale files, now I'm
not so sure.
* feat: Added simple stack trace output
* feat: Added backtrace wrapper
* build: Added libbacktrace include dirs to build
* build: Fixed libbacktrace variables
* build: More backtrace variable fixes
* build: Try to find backtrace include
* build: Get backtrace header from cmake script
* feat: Added backtrace support for execinfo
* feat: Added support for StackWalk on Windows
* remove flatpak
rationale: if we need it again we can always get it back from old commits
* remove double spaces
* remove Fedora rawhide
* remove Windows installer signing
* separate Ubuntu and AppImage builds
* clear project context when closing all providers
* Show project name on window title
* refactor RequestChangeWindowTitle to RequestUpdateWindowTitle
* add spaces
* refresh store on open
* show the user if the request failed
* do not refresh if a refresh is in progress
This makes ImHex crash
* InProgress
* fix formatting
* use a localized message
* set localized message in other langs
* Tell the user the net request which failed
* Show signal name
* Print execution time of startup tasks
* replace NULL with nullptr
* change wording for task logs
* Remove '..' from task name
* remove using namespace chrono
* use action for ccache + limit cache size
* remove 'prepare cache' step for Windows
* remove libpython from CI
* fix the restore key for MacOS
* remove Rust from CI
* do not pin appimage-builder version since issue with it has been resolved
* use ccache action for CodeQL
* Rebased onto latest master
Co-authored-by: WerWolv <werwolv98@gmail.com>
* add error message for the file picker
* set a different message for Windows/MacOS
* fix typos
* Moved file dialog error handling to a callback function
Co-authored-by: Nik <werwolv98@gmail.com>
* added update information
* upload zsync
* set version of appimage
* release: do not execute steps which needs a token when not provided
* set the appimage output name directly in the recipe file
* add metainfo file in every Linux package
* update metainfo file
* set categories in the metainfo file
* add both .metainfo.xml and .appdata.xml files because it is the old standard still wanted by some things
* actually use WerWolv's repository for AppImage updates
* rename metainfo files
* Typo : appdata and not appinfo
* split screenshots
* Added option to bundle CA
* use bundled CA for AppImage
* Fix bundled CA not working on Linux
* revert change to add null terminated string
* set IMHEX_USE_BUNDLED_CA to ON on Windows
* install deb package in a different folder than AppImage
* added comment for Ubuntu cmake build
* fixed typos
* separate cmake build for deb and appimage
* feat: Initial implementation of an Intel Hex provider
* fix: Reading of bytes from intel hex files
* lang: Added localization for new provider
* ui: Only show file name in intel hex provider name
* feat: Added Motorola SREC provider
This avoids a crash on POSIX after running a command from the command
pallete; on completion, a SIGCHILD will be raised, which was incorrectly
calling the error handler and terminating ImHex.
Co-authored-by: Eoin Mcloughlin <helloworld@eoinrul.es>
* sys: Initial effort to replace existing project files with a better system
* sys: Added back marking provider as dirty
* sys: Remove git commit information from project files
* sys: Format data processor save file nicely
* fix: Automatic pattern loading not working correctly
* ui: Added warning popup when closing a provider with modifications
Closes#604
* sys: Fixed build issues
* tests: Removed useless debug logs
* patterns: Updated pattern language
* sys: Added log message when crashing with a signal
* sys: Make sure abnormal termination handlers are being called more reliably
* Update ja_JP.cpp
- Corrected some mistranslations
- Updated some outdated class names
- Added some translations
- Replaced all exclamation marks (！) with Japanese reading marks (。)
- Replaced all 3-dot readers (...) with 2-byte character versions (…)
* Update ja_JP.cpp
Corrected mistranslation of "Contributors"
* RUST_PATH doesn't seem to be used on ArchLinux
* renamed jobs
* no objc code on Windows (need confirmation)
* fakeroot strace patchelf for Ubuntu : not needed ?
* try to use some system libraries on Arch
* added comments to scroll to cmake instruction easily
* No ojbC code on Ubuntu
* cmake: use GNUInstallDirs to find install dirs on Linux
* install plugins to lib/imhex/plugins
* fix included files in imhex.spec
* fix the release CI + do not upload x86_64 folder for Fedora
* change rpm names
* added IMHEX_STRIP_RELEASE option to optionally strip releases files (was done all the time before)
* Customize our imhex.spec file (use online building for our Fedora package)
* added IMHEX_PLUGINS_IN_SHARE option for AppImage
* test
* chore: bug report template readability fix
Signed-off-by: Ryan Russell <git@ryanrussell.org>
* refactor(imgui-rs): Doc aliases readability fixes
Signed-off-by: Ryan Russell <git@ryanrussell.org>
* refactor(imgui-rs): Code comment readability
Signed-off-by: Ryan Russell <git@ryanrussell.org>
* feat: Added LEB128 in data inspector
* feat: Added support for editing LEB128 values
* Moved LEB functions from utils.cpp to crypto.cpp
* Added placeholders for translations
* Made DataInspector::impl::Entry.maxSize mandatory
* Fixed undefined leftshifting behaviour
* Ci: use matrix for Fedora
* Ci: use ccache in Fedora build
* Ci: cache CMakeCache for Fedora
* build type release fedora
* Rawhide with lowercase
* strip binaries on Linux
* strip files for CPack (didn't test)
* add soversion to libimhex on Linux and MacOS
* fixed that typo
fix not saving to "XDG_DATA_HOME", when "XDG_DATA_DIRS" is available.
fix "Content Store" persist in flatpak.
XDG specification specifies how to find config and data directories on
linux systems. Specifically, it says this:
- Data should be written to $XDG_DATA_HOME
- Config should be written to $XDG_CONFIG_HOME
- Data should be read from $XDG_DATA_HOME:$XDG_DATA_DIRS
- Config should be read from $XDG_CONFIG_HOME:$XDG_CONFIG_DIRS
* store version in file
* use version file in release workflow
* use new version file in build workflow
* ArchLinux build
* setup cache for ArchLinux
* add version check in release CI
* edit step description
* update pkgbuild to install correctly
* AUR deploy
* rename version file to VERSION
* install all default plugins in PKGBUILD
* Added emojis to build workflow
* Added emojis to release workflow
* separate update packages and install dependencies in two steps
* fix Release CI
* add md5Sums to PKGBUILD
* make PKGBUILD point to the official repo + set v in tag
* Version file test
* change artifacts names
* Release updated
* Set release job name
* set branch to the branch used to release
* Set version for Windows Portable
* git: New cache update (#558)
* add CMakeCache.txt to cache
* added workflow_dispatch to workflows
* remove restore-keys from workflows
* Separated cache
* re-added restore-keys
* put the version file for all OS
* fix: In/Out parameters not working correctly when using the preprocessor
* update submodules
Co-authored-by: WerWolv <werwolv98@gmail.com>
* build: Experimentally switch to gcc on macOS
* build: Corrected gcc paths
* build: Enable objective c support on macOS
* build: Enable ObjC and ObjC++ on macOS
* build: Add ObjC and ObjC++ flags
* build: Try compiling objc with clang
* build: Remove invalid flags again
* fix: Let's not include objc headers in C++ code
* sys: Move macos utils code to its own file
* fix: Missing unistd include on mac
* sys: Removed loader script stuff since it's currently unused and broken
* fix: Missing include
* fix: Another missing include
* fix: CFURLCreateWithBytes wants a pointer to mutable data
* fix: Try disabling name mangling of ObjC functions
* sys: Move macos utils declarations to its own header file
* fix: C Linkage
* fix: Move objc function prototypes to C++ headers
* fix: More missing includes
* fix: Warning error
* sys: Call ObjC with C ABI instead of trying to use C++
* build: Update libraries
* sys: Fixed build errors
* sys: No const correctness I guess
* sys: Fixed prototypes
* sys: This is C now
* sys: More nullptr -> NULL
* sys: Fix crash on exit
* sys: Try using proper std concepts instead of custom ones
* sys: Replaced another hex::is_signed
* build: Upgrade to gcc 12 and MacOS Monterey
* build: Fixed MacOS runner name
* build: Cache correct ccache folder on macOS
* build: Initial attempt to fix .dmg mess
* build: Still download database resources on other OSes
* build: Try fixing path recursion error
* build: Move main executable and database files to correct folder
* build: Install main executable without installing python stuff
* build: Move things around to maybe get them to be bundled
* I wanna die
* renamed imhex.app to Imhex.app
* net.WerWolv.ImHex
Co-authored-by: WerWolv <werwolv98@gmail.com>
* Restyled macOS icon for Big Sur
* Update other icons
* Different gradients
* Reverted macOS icon because of design guidelines
* Final version of new icons
* ui/ux: Initial recreation of the hex editor view
* ui/ux: Added back support for editing cells
* ux: Make scrolling and selecting bytes feel nice again
* ui/ux: Improved byte selecting, added footer
* sys: Make math evaluator more generic to support integer only calculations
* patterns: Moved value formatting into pattern language
* ui/ux: Added Goto and Search popups, improved selection
* ui: Added better tooltips for bookmarks and patterns
* sys: Use worse hex search algorithm on macOS
Sadly it still doesn't support `std::boyer_moore_horsepool_searcher`
* ui: Added back missing events, menu items and shortcuts
* fix: Bookmark highlighting being rendered off by one
* fix: Various macOS build errors
* fix: size_t is not u64 on macos
* fix: std::fmod and std::pow not working with integer types on macos
* fix: Missing semicolons
* sys: Added proper integer pow function
* ui: Added back support for custom encodings
* fix: Editor not jumping to selection when selection gets changed
* ui: Turn Hexii setting into a data visualizer
* sys: Added back remaining shortcuts
* sys: Remove old hex editor files
* sys: Moved more legacy things away from the hex editor view, updated localization
* fix: Hex editor scrolling behaving weirdly and inconsistently
* sys: Cleaned up Hex editor code
* sys: Added selection color setting, localized all new settings
* fix: Search feature not working correctly
* ui: Replace custom ImGui::Disabled function with native ImGui ones
* ui: Fix bookmark tooltip rendering issues
* fix: Another size_t not being 64 bit issue on MacOS
* sys: Make ImHex compile with -Wall -Wextra -Werror
* sys: Fixed various build errors on Linux
* sys: Explicitly ignore return value of `system` function
* sys: More fixes for the warnings GitHub Actions enables somehow
* sys: More fixes
* sys: Remove -Werror again to see all GitHub Actions warnings
* sys: Hopefully fixed all remaining warnings
* sys: Added back -Werror
* git: Change windows icon in GitHub Actions
* refactor(patterns): add visitor interface
* refactor(patterns): add public accessors
* refactor(patterns): add method to get pattern value
* refactor(pattern): make some methods public
* refactor(pattern): extract code to draw GUI
* refactor(patterns): remove GUI related code from patterns
* refactor: move common GUI function from pattern to pattern_drawer
* refactor(pattern_drawer): extract common code into methods
* refactor: rename ImGuiDrawer -> PatternDrawer
* refactor(patternr): move displayEnd into PatternDrawer
* refactor: use ArrayPattern concept to restrict argument type
Co-authored-by: Dmitry Polshakov <dmitry.polshakov@dsr-corporation.com>
* build: Tried to streamline and fix AppImage building
* build: Tried fixing build
* build: Added back version key
* build: Fixed AppImage name
* build: Fixed python bundling, fixed ELF uploading
* build: Don't upload a bunch of AppImage garbage files in the portable ELF version
* build: Fixed another wrong library path
* build: Removed old manual AppImage building scripts
* fix(eval): restore data offset for local variables
* test(pattern_lang): add tests for arrays, nested structs and functions
Co-authored-by: Dmitry Polshakov <dmitry.polshakov@dsr-corporation.com>
* patterns: Initial work to refactor pattern language to use smart pointers
* patterns: Fixed remaining issues, moved patterns to unique files
* sys: Added missing includes for macOS
clipper.Step() should be called in a loop until it returns false.
You don't need to call End() after it.
Co-authored-by: Dmitry Polshakov <dmitry.polshakov@dsr-corporation.com>
* fix: scan include name until end of line
* fix: correctly check file existance
Co-authored-by: Dmitry Polshakov <dmitry.polshakov@dsr-corporation.com>
* fix: NodeBufferSlice bounds validation
* fix: Make sure buffer is within u64 bounds in NodeCastBufferToInteger
* nodes: Use specific output methods in number constants
* sys: Initial refactoring of the SharedData class
* sys/pattern: More refactoring, make every provider have its own patterns
* sys: Finished up refactoring. No more SharedData!
* sys: Fixed compile on Unix
* tests: Fixed unit tests
* sys: Moved view and lang files
* pattern: Added assignment operator support to for loops
* tests: Fixed compile issue
* Simplify creating an AppImage
- package.sh packages the build result into an AppImage. It requires
squashfs-tools to work
- runtime-x86_64 is a binary distributed by the AppImage project that takes
care of extracting and running the image
* use AppRun from AppImage project
* clean up no longer needed bits
* Keep docker way of working around
- Docker now also uses `package.sh`, so no more need for FUSE.
- Fetch binaries instead of storing in git.
* Fix details
* wait a little longer for the container to start
* Update TEST_ASSERT to do nothing if condition is true
The TEST_ASSERT should not return if the condition is true, because:
- it prevents the usage of multiple TEST_ASSERT in a single test case,
- that behavior differs from how the assert in the standard library
works, and thus may give unexpected results.
Make the TEST_ASSERT to print an error message (with an formatted
optional user part) when it fails to make debugging easier.
* Fix some bugs in TestProvider, add unit tests
Use pointer-to-vector in TestProvider so writes can be tested, too.
* Add test EncodeDecode16, fix some encode16 bugs
The function mbedtls_mpi_write_string needs a bit longer buffer than the
resulting string actually will be.
Known bug: mbedtls_mpi_read_binary ingores initial null bytes
* Add test EncodeDecode64, fix some bugs
The functions mbedtls_base64_encode and mbedtls_base64_decode needs a
bit longer buffer than the resulting string actually will be.
* Remove check for empty data from TestProvider
It can be valid to get the hash of empty string.
* Add tests for CRC calculation
Two type of thests:
- compare the result of the CRC calculation to a known to be good
results,
- generate random data as message, calculate of it's CRC and append that
to the message, the CRC of this new data should be 0.
* Add test for hash algorithms
* Add includes in tests
* Remove the use of C++20 ranges
It seems that Apple Clang does not support range-based constrained
algorithms at this time.
* Replace encode16 implementation
To encode the zero bytes at the begining of the input vector, too.
* Filter by regex in string view
* Dont recompile the regex for every string, display error message
* localization
* Use data->Buf for pattern creation / searching
The filter string seems to get updated after the callback finished.
Therefore the search string was always 1 character behind the actual
string in the textfield when calling find() / creating the regex.
* Fix CRC calculation, add more CRC parameters
Use the Boost CRC module to calculate the CRC values.
Add options for final xor value, reflectIn and reflectOut.
Fixes#320
* Cleanup Hash view combo box, add CRC8
* Use offset/size consistently
* Cleanup: unify processing data by chunks
* Change CRC algorithm back, drop boost dependency
This is mostly the original algorithm, with a few fixes and small
additions (support for reflect In / Out, final XOR value).
* Use size_t for file read size consistently
* sys: Updated curl to latest version
* sys: Fix macOS compilation
* ui: Fix splash screen OpenGL init for macOS
* sys: Fix std::min compile errors
* git: Re-enabled macos workflow
* sys: Remove includes of the range library
* build: Find OpenGL using CMake
* sys/build: Fix bundled plugins on macOS
* build: Copy plugins to bundle when creating a bundle
* build: Fixup bundled plugins
* sys: Search for plugins in the bundle instead of in Application Support
* sys: Allow resources to be placed in multiple directories on macOS
* build: Output built plugins to the plugins/ directory when not creating a bundle on macOS
* sys: Fix Application Support paths on macOS
* sys: Define ftruncate64 on macOS
* sys: Fix absolute value computation for std::string::at on macOS
Co-authored-by: WerWolv <werwolv98@gmail.com>
* fix: crashes when connecting with no COM port
* feat: remove `View` suffix for view menu
* i18n: add more to Chinese(Simplified)
* Revert "Merge branch 'master' of github.com:xtexChooser/ImHex"
This reverts commit 8afcfe8f9d, reversing
changes made to 7651ad6661.
* patterns: Rewrite most of the evaluator to mainly use polymorphism instead of just RTTI
* patterns: Fixed a couple of AST memory leaks
* patterns: Parse string operations correctly
* patterns: Various fixes and cleanup
* patterns: Implement primitive function definitions
Function parameters now need to provide their type in the definition
* patterns: Added function variable definition and assignment
* patterns: Added remaining function statements
* patterns: Added unsized and while-sized arrays
* patterns: Added multi variable declarations to functions
* patterns: Added std::format built-in function
* patterns: Allow passing custom types to functions
* patterns: Added attributes and new "format" attribute
* patterns: Use libfmt for std::print instead of custom version
* patterns: Remove unnecessary string compare function
* pattern: Fix preprocessor directives
* patterns: Fix unit tests
* patterns: Added cast expression
* patterns: Handle endianess in function parameters
* patterns: Added casting to different endian
* patterns: Added 'str' type for functions
* feat(i18n): add Chinese(Simplified) translations
* feat: keep console window on Windows for debug builds
* feat(18n)
* feat(i18n): improve Chinese translation
* fix: unify the line terminators
* feat(build): formatting
* fix: exclude from SysWow64 for 64bits windows
* Revert "fix: exclude from SysWow64 for 64bits windows"
This reverts commit a6d66a4a56.
* Revert "feat: keep console window on Windows for debug builds"
This reverts commit 9fd4699c9f.
- std::string -> const std::string& where needed
- Added a FileIO abstraction class
- Fixed recent files not updating
- Removed localization file from global include
- Renamed lang to pattern_language/pl
- Renamed EventFileDropped to RequestFileOpen
* removing superfluous slash from paths
* Add a necessary rpath to the imhex binary
* add a little error handling to dlopen
* fall back on en-US if no language specified in prefs
* PR changes as per @WerWolv
* build: Add libcurl
* build: Stop the whole static linking on Windows mess. There's no use anymore
* sys: Added update detector and moved some startup tasks to splash screen
* sys: Updated ImHex icon
* fix: Settings button on welcome page once again works
* build: Fix build on Linux
* sys: Fixed splash window not searching all paths for resources
getSetting now returns a straight nlohmann::json instead of an optional.
If the data isn't present, it will return a json null.
All accesses to the settings will first check that the data has the
expected type.
This change is to reflect the commit made in 785ecb8a78, which changes openssl libcrypto to libmbedtls, making changes everywhere except in the README documentation.
* sys: Move away from metadata paths next to executable in the application
Build system doesn't properly install / pack stuff yet
* build: Updated README to contain better install instructions
* sys: Search for imhex resource files in ~/Application Support
* sys: MAX_PATH -> PATH_MAX
* sys: Seach for imhex resource files in Application Support using NSFileManager (#180)
* sys: Allow for multiple file search paths
Also use install prefix instead of just /usr on Linux
* build: Fixed IMHEX_INSTALL_PREFIX macro definition
* build: Fix duplicate switch entry on Linux
* docs: Updated readme to properly reflect new paths and dependencies
* sys: Install files in their proper paths on linux (#183)
* Install files in their proper paths on linux
* Only create user directories
* Follow the XDG specification on linux
XDG specification specifies how to find config and data directories on
linux systems. Specifically, it says this:
- Data should be written to $XDG_DATA_HOME
- Config should be written to $XDG_CONFIG_HOME
- Data should be read from $XDG_DATA_HOME:$XDG_DATA_DIRS
- Config should be read from $XDG_CONFIG_HOME:$XDG_CONFIG_DIRS
The default values are this:
- XDG_DATA_HOME: $HOME/.local/share
- XDG_CONFIG_HOME: $HOME/.config
- XDG_DATA_DIRS: /usr/share:/usr/local/share
- XDG_CONFIG_DIRS: /etc/xdg
Platforms with non-standard filesystems (like NixOS) will correctly set
up those environment variables, allowing softwares to work unmodified.
In order to make integration as simple as possible, we use a simple
header-only dependency called XDGPP which does all the hard work for us
to find the default directories.
* Look for plugins in all Plugin Paths
If the plugin folder was missing from one of the PluginPaths, we would
immediately stop loading plugins. We now keep looking even if one of the
path is missing.
Co-authored-by: Nichole Mattera <me@nicholemattera.com>
Co-authored-by: Robin Lambertz <unfiltered@roblab.la>
* dont dlclose nullptrs in plugins
this fixes a segmentation fault at startup when dlclose is called with a nullptr
* Style fixes
Co-authored-by: WerWolv <werwolv98@gmail.com>
* Properly bundle all files in linux install step
* Run make install on linux build
* Use correct default magic database
* Set default magic db permissions correctly
* Fixed magic file detection issues
* Don't install default magic file if none was found
* Try fix windows packing issues
* Added imnodes
* Added basic data processor view. Still needs to be cleaned up
* Make sure all attached links get properly removed when a Node is deleted
* Cleanup and API exposing
* Added data provider overlays and integrate them with the data processor
* Optimized data processing
* Node UI enhancements
* Added support for all themes to the nodes editor
* Improved data processor context menus
* Fixed data processor context menu showing up everywhere
* Make hex editor context menu behave the same as data processor one
* Add different node pin types and prevent incompatible ones from being connected
* Don't require explicitly marking node as end node
* Fixed plugin copying
* Added some more nodes
* Create ImHex WIX Installer
* Build Windows Installer in github actions
* Have a single windows step handling MSI and ZIP.
* Properly copy dependencies to portable zip
* Add magicdb to zip/msi
* Use arial font and smaller text in license
* Initial parser rewrite effort
Lexer and Token cleanup, Parser started over
* Greatly improved parser syntax
* Reimplemented using declarations and variable placement parsing
* Added back unions and structs
* Added enums as well as mathematical expressions (+, -, *, /, <<, >>, &, |, ^)
* Code style improvement
* Implemented arrays and fixed memory issues
* Fixed more memory issues in parser, reimplemented validator, evaluator and patterns
* Fixed builtin types, arrays and reimplemented strings
* Improved error messages
* Made character a distinct type, used for chars and strings
* Implemented padding, fixed arrays
* Added bitfields
* Added rvalue parsing, no evaluating yet
* Added .idea folder to gitignore
* Fixed build on MacOS
* Added custom implementation of integral concept if not available
* Rebased onto master
* Fixed array variable decl crash
* Added rvalues and dot syntax
* Lower case all pattern language error messages
* Fixed typo in variable name
* Fixed bug where preprocessor would not ignore commented out directives
* Reimplemented pointers
* Fixed rebase issues
* Build refactoring and initial plugin support
* Possibly fixed linux / mac build
* Added libdl to libglad build script
* Add glfw to imgui dependencies
* Refactored common functionality into "libimhex" for plugins
* Added plugin loading and example plugin
* Added proper API for creating a custom view and a custom tools entry with plugins
* add glm to arch deps
After running got `None of the required 'glm' found`. This fixes that
* dist/fedora: Include file magic headers
Due to differences in package names between Deb based systems, Arch
Linux, and RPM based systems the package containing the development
headers for file were missing from the Fedora dependencies script.
This includes the package `file-devel`, which is the package which
resolves the issue.
In Fedora, one can identify the package providing a specific file using
the verb "whatprovides" with the command dnf, e.g.:
[~]$ dnf whatprovides /usr/include/magic.h
Last metadata expiration check: 4 days, 0:23:05 ago on Fri 04 Dec 2020 09:06:53 AM PST.
file-devel-5.39-3.fc33.i686 : Libraries and header files for file development
Repo : fedora
Matched from:
Filename : /usr/include/magic.h
file-devel-5.39-3.fc33.x86_64 : Libraries and header files for file development
Repo : @System
Matched from:
Filename : /usr/include/magic.h
file-devel-5.39-3.fc33.x86_64 : Libraries and header files for file development
Repo : fedora
Matched from:
Filename : /usr/include/magic.h
If one is unsure of the specific path, globbing may be used (but must be
quoted):
dnf whatprovides "*/magic.h"
Resolves#48
* dist: Prevent already installed packages in ArchLinux and MSYS2.
Use --needed option with pacman to prevent it.
* Add script to install dependencies on Debian/Ubuntu.
Tested with Xubuntu 20.04 and Debian testing
(in today's Docker image bitnami/minideb).
Update README.md.
* ci: rework (#31)
* Support non standard LLVM library names (#86)
This fix openSUSE and Gentoo issue mentioned in https://github.com/WerWolv/ImHex/issues/37#issuecomment-739503138.
(tested on openSUSE tumbleweed via Docker)
I also took the liberty of renaming llvm_lib to llvm_demangle_lib to be more specific in the ``CMakeLists.txt``.
* Implement proper DPI handling
* Implement basic custom font support
* Fix building on windows
* Hopefully fix fonts on Windows
* Fix several scaling issues
* Replace font renderer with freetype
* Updated CI and dependency scripts
* Rebuild default font atlas
* Correct platform detection macro for mingw
* Fixed PKGBUILD
Co-authored-by: brockelmore <31553173+brockelmore@users.noreply.github.com>
Co-authored-by: Brian 'Redbeard' Harrington <redbeard@dead-city.org>
Co-authored-by: Biswapriyo Nath <nathbappai@gmail.com>
Co-authored-by: Stéphane Gourichon <stephane.gourichon@fidergo.fr>
Co-authored-by: umarcor <38422348+umarcor@users.noreply.github.com>
Co-authored-by: Mary <me@thog.eu>
Co-authored-by: WerWolv <werwolv98@gmail.com>
This fix openSUSE and Gentoo issue mentioned in https://github.com/WerWolv/ImHex/issues/37#issuecomment-739503138.
(tested on openSUSE tumbleweed via Docker)
I also took the liberty of renaming llvm_lib to llvm_demangle_lib to be more specific in the ``CMakeLists.txt``.
Due to differences in package names between Deb based systems, Arch
Linux, and RPM based systems the package containing the development
headers for file were missing from the Fedora dependencies script.
This includes the package `file-devel`, which is the package which
resolves the issue.
In Fedora, one can identify the package providing a specific file using
the verb "whatprovides" with the command dnf, e.g.:
[~]$ dnf whatprovides /usr/include/magic.h
Last metadata expiration check: 4 days, 0:23:05 ago on Fri 04 Dec 2020 09:06:53 AM PST.
file-devel-5.39-3.fc33.i686 : Libraries and header files for file development
Repo : fedora
Matched from:
Filename : /usr/include/magic.h
file-devel-5.39-3.fc33.x86_64 : Libraries and header files for file development
Repo : @System
Matched from:
Filename : /usr/include/magic.h
file-devel-5.39-3.fc33.x86_64 : Libraries and header files for file development
Repo : fedora
Matched from:
Filename : /usr/include/magic.h
If one is unsure of the specific path, globbing may be used (but must be
quoted):
dnf whatprovides "*/magic.h"
Resolves#48
This document is a guide for developers who want to contribute to ImHex in any way. It contains information about the codebase, the build process and the general workflow.
## Making Changes
### Adding new features
If you'd like to add new features, the best way to start is by joining our Discord and telling us about your idea. We can then discuss the best way to implement it and how it should be integrated into ImHex or if it should be done in a separate plugin.
There are standalone plugin templates that use ImHex as a submodule. You can find them here:
If you'd like to support a new language in ImHex, the best way is by using the `dist/langtool.py` tool. It will create the necessary file for you and help you fill them out.
First, run the tool with `python3 dist/langtool.py create plugins/builtin/romfs/lang <iso_code>` where `<iso_code>` is the ISO 639-1 code of your language. This will create a new file in the language directory.
Afterwards follow the prompts of the program to populate the entire file. Once you're done, rerun cmake and rebuild ImHex. Your language should now be available in the settings.
### Updating an existing language
If you'd like to add missing keys to an existing language, you can also use the `dist/langtool.py` tool. Run it with `python3 dist/langtool.py translate plugins/builtin/romfs/lang <iso_code>` where `<iso_code>` is the ISO 639-1 code of the language.
This will one by one list all the missing translation keys that are present in the default translation file, and you can fill them in with the correct translation for your language.
## Codebase
ImHex is written in C++ and usually uses the latest compiler and standard library features available in gcc on all supported OSes. At the time of writing this is C++23.
### Structure
-`main`: Contains the main application code
- Important to understand here is that the main ImHex application is basically just an empty shell.
- All it does is create a Window and a OpenGL context using GLFW, load all available plugins, properly configure ImGui and render it to the screen.
- Everything else is done inside of plugins. ImHex comes with a few plugins by default, most notably the `builtin` plugin which contains the majority of the application code.
- In most cases, this code doesn't need to be modified. Most features should be self-contained inside a plugin.
-`lib`
-`libimhex`: Contains all helper utilities as well as various APIs for plugins to interact with ImHex.
- The library's main purpose is for Dependency Inversion. The ImHex main application as well as libimhex do not know about the existence of plugins at build time. Plugins and the main application instead link against libimhex and use it as a common API to interact with each other.
- Since libimhex is a doesn't know about the existence of plugins, it cannot depend on any of them. This includes localizations and things that get registered by plugins after launch.
- Even if the builtin plugin is technically always available, it is still a plugin and should be treated that way.
- All important APIs can be found in the `hex/api` include directory and are documented in the respective header file.
-`external`: All libraries that need custom patches or aren't typically available in package managers go into here.
- If you'd like to add new features to the Pattern language, please make a PR to https://github.com/WerWolv/PatternLanguage instead. ImHex usually depends on the latest commit of the master branch of this repo.
-`plugins`
-`builtin`: The builtin plugin. Contains the majority of the application code.
- It's the heart of ImHex's functionality. It contains most of the default views, providers, etc. so if you want to add new functionality to ImHex, this is the place to start.
-`tests`: Contains all unit tests for ImHex. These are run automatically by the CI and should be kept up to date, especially when adding new helper functions to libimhex.
### RomFS
ImHex uses a custom library called [libromfs](https://github.com/WerWolv/libromfs). It's a simple static library which uses CMake's code generation feature to embed files into the binary at compile time so they can be accessed at runtime.
All plugins have a `romfs` folder which contains all files that should be embedded into the binary. Resources that need to be embedded into the main application (this is usually not necessary), go into the `resources/romfs` folder.
When adding, changing files or removing files, make sure to re-run CMake to update the generated code. Otherwise, the changes might not be reflected in the binary.
## Development Environment
I personally use CLion for development since it makes configuring and building the project very easy on all platforms.
### Windows
- Install MSYS2 from https://www.msys2.org/ and use the `dist/get_deps_msys2.sh` script to install all dependencies.
### Linux
- Install all dependencies using one of the `dist/get_deps_*.sh` scripts depending on your distribution or install them manually with your package manager.
### macOS
- Install all dependencies using brew and the `dist/Brewfile` script.
The easiest way to install ImHex is to download the latest release from the [GitHub Releases page](https://github.com/WerWolv/ImHex/releases/latest).
There's also a NoGPU version available for users who don't have a GPU or want to run ImHex in a VM without GPU passthrough.
### Windows
#### Installer
Simply run the installer to install ImHex on your system
#### Portable
Extract the zip file to any location on your system.
### macOS
Simply use the drag-n-drop dmg package to install ImHex on your system. It's possible that you need to allow the app to run in the security settings.
### Linux
#### AppImage
To run the AppImage, make sure it's executable. Then simply run it.
```bash
chmod +x imhex-*.AppImage
./imhex-*.AppImage
```
If you're experiencing glib / libgtk assertion failures, you might need to setup your `XDG_DATA_DIRS` env var correctly. In this case, run the following command before executing the AppImage. (See issue [ImHex/#2038](https://github.com/WerWolv/ImHex/issues/2038))
```bash
exportXDG_DATA_DIRS="/usr/local/share:/usr/share"
```
#### Flatpak
To install the Flatpak, make sure you have the Flathub repository added to your system. Then simply run the following command:
```bash
flatpak install flathub net.werwolv.ImHex
```
#### Ubuntu DEB Package
To install the DEB package, simply run the following command:
```bash
sudo apt install ./imhex-*.deb
```
#### Arch Linux
To install the Arch Linux package, simply run the following command:
```bash
sudo pacman -U imhex-*.pkg.tar.zst
```
#### Fedora / RHEL / AlmaLinux RPM Package
To install the RPM package, simply run the following command:
```bash
sudo dnf install ./imhex-*.rpm
```
## Nightly Builds
The GitHub Actions CI builds a new release package on every commit made to repository. These builds are available on the [GitHub Actions page](https://github.com/WerWolv/ImHex/actions?query=workflow%3A%22Build%22).
These builds are not guaranteed to be stable and may contain bugs, however they also contain new features that are not yet available in the official releases.
## Building from source
Build instructions for Windows, Linux and macOS can be found under `/dist/compiling`:
- Windows: [Link](dist/compiling/windows.md)
- macOS: [Link](dist/compiling/macos.md)
- Linux: [Link](dist/compiling/linux.md)
## Package managers
ImHex is also available on various package managers. The officially supported ones are listed here:
### Windows
- **Chocolatey**
- [imhex](https://community.chocolatey.org/packages/imhex) (Thanks to @Jarcho)
ImHex is entirely built around the possibility to easily load plugins (most of it's features are actually implemented as a plugin!).
To install plugins, simply download the relevant `.hexplug` file and drop it in your `plugins` folder. The location of that folder can be found under `Help -> About -> ImHex Directories`.
## Available Plugins
(If you're developing a Plugin on your own, please feel free to add it to this list)
<a title="Download the latest version of ImHex" href="https://imhex.download"><img alt="Download the latest version of ImHex!" src="resources/dist/common/get_release_banner.png"></a>
<a title="Download the latest nightly pre-release version of ImHex" href="https://imhex.download/#nightly"><img alt="Download the latest nightly pre-release version of ImHex" src="resources/dist/common/get_nightly_banner.png"></a>
<a title="Use the Web version of ImHex right in your browser!" href="https://web.imhex.werwolv.net"><img alt="Use the Web version of ImHex right in your browser!" src="resources/dist/common/try_online_banner.png"></a>
<a title="Read the documentation of ImHex!" href="https://docs.werwolv.net"><img alt="Read the documentation of ImHex!" src="resources/dist/common/read_docs_banner.png"></a>
</p>
## Supporting
If you like my work, consider supporting me on GitHub Sponsors, Patreon or PayPal. Thanks a lot!
If you like my work, please consider supporting me on GitHub Sponsors, Patreon or PayPal. Thanks a lot!
<summary><strong>Easy to get started</strong></summary>
## Additional Files
- Support for many different languages
- Simplified mode for beginners
- Extensive documentation
- Many example files available on [the Database](https://github.com/WerWolv/ImHex-Patterns)
- Achievements guiding you through the features of ImHex
- Interactive tutorials
</details>
For format patterns, includable libraries and magic files, check of the [ImHex-Patterns](https://github.com/WerWolv/ImHex-Patterns) repository here. Feel free to PR your own files there as well!
## Pattern Language
The Pattern Language is the completely custom programming language developed for ImHex.
It allows you to define structures and data types in a C-like syntax and then use them to parse and highlight a file's content.
For format patterns, libraries, magic and constant files, check out the [ImHex-Patterns](https://github.com/WerWolv/ImHex-Patterns) repository.
**Feel free to PR your own files there as well!**
## Requirements
To use ImHex, the following minimal system requirements need to be met.
> [!IMPORTANT]
> ImHex requires a GPU with OpenGL 3.0 support in general.
> There are releases available (with the `-NoGPU` suffix) that are software rendered and don't require a GPU, however these can be a lot slower than the GPU accelerated versions.
>
> If possible at all, make ImHex use the dedicated GPU on your system instead of the integrated one.
> ImHex will usually run fine with integrated GPUs as well but certain Intel HD GPU drivers on Windows are known to cause graphical artifacts.
- **OS**:
- **Windows**: Windows 7 or higher (Windows 10/11 recommended)
- **macOS**: macOS 13 (Ventura) or higher,
- Lower versions should still work too, but you'll need to compile ImHex yourself. The release binaries will NOT work.
- The macOS build is not signed and will require you to manually allow them in the Security & Privacy settings.
- **Linux**: "Modern" Linux. The following distributions have official releases available. Other distros are supported through the AppImage and Flatpak releases.
- Ubuntu and Debian
- Fedora
- RHEL/AlmaLinux
- Arch Linux
- Basically any other distro will work as well when compiling ImHex from sources.
- **CPU**: Officially supported are x86_64 and ARM64, though any Little Endian 64 bit CPU should work.
- **GPU**: OpenGL 3.0 or higher
- Integrated Intel HD iGPUs are supported, however certain drivers are known to cause various graphical artifacts, especially on Windows. Use at your own risk.
- In case you don't have a GPU available, there are software rendered releases available for Windows and macOS
- **RAM**: ~150MiB, more is required for more complex analysis
- **Storage**: 150MiB
## Installing
Information on how to install ImHex can be found in the [Install](/INSTALL.md) guide
## Compiling
This guide assumes you're either on Windows using mingw or on Arch Linux
To compile ImHex on any platform, GCC (or Clang) is required with a version that supports C++23 or higher.
Windows and Linux releases are being built using latest available GCC.
MacOS releases are being built using latest available LLVM Clang.
You need a C++20 compatible compiler such as GCC 10.2.0 to compile ImHex.
The following libraries are needed to compile ImHex. All of them can be found in the default pacman repositories
Important to note is, the MSVC and AppleClang compilers are both **NOT** supported since they're both generally severely outdated and lack features GCC and LLVM Clang have.
All in one: sudo pacman -S glfw file openssl capstone llvm llvm-libs nlohmann-json python3
All in one for mingw: pacman -S mingw-w64-x86_64-gcc mingw-w64-x86_64-cmake mingw-w64-x86_64-make mingw-w64-x86_64-capstone mingw-w64-x86_64-glfw mingw-w64-x86_64-glm mingw-w64-x86_64-file mingw-w64-x86_64-llvm mingw-w64-x86_64-nlohmann-json mingw-w64-x86_64-openssl mingw-w64-x86_64-polly mingw-w64-x86_64-python
```
> [!NOTE]
> Many dependencies are bundled into the repository using submodules so make sure to clone it using the `--recurse-submodules` option.
> All dependencies that aren't bundled, can be installed using the dependency installer scripts found in the `/dist` folder.
After all the libraries are installed, run the following commands to build ImHex
```
mkdir build
cd build
cmake ..
make -j
```
For more information, check out the [Compiling](/dist/compiling) guide.
On Windows, download the python standard library from https://github.com/python/cpython/tree/master/Lib and place the files and folders in `lib/python3.8` next to your built executable. Don't forget to also copy the `libpython3.8.dll` and `libwinpthread-1.dll` from your mingw setup next to the executable.
## Contributing
See [Contributing](/CONTRIBUTING.md)
## Plugin development
To develop plugins for ImHex, use the following template project to get started. You then have access to the entirety of libimhex as well as the ImHex API and the Content Registry to interact with ImHex or to add new content.
On both Windows and Linux, copy the files from `python_libs` in the `lib` folder next to your built executable.
Place your magic databases in the `magic` folder next to your built executable
Place your patterns in the `pattern` folder next to your built executable
Place your include pattern files in the `include` folder next to your built executable
## Credits
- Thanks a lot to ocornut for their amazing [Dear ImGui](https://github.com/ocornut/imgui) which is used for building the entire interface
- Thanks to orconut as well for their hex editor view used as base for this project.
- Thanks to BalazsJako for their incredible [ImGuiColorTextEdit](https://github.com/BalazsJako/ImGuiColorTextEdit) used for the pattern language syntax highlighting
- Thanks to AirGuanZ for their amazing [imgui-filebrowser](https://github.com/AirGuanZ/imgui-filebrowser) used for loading and saving files
- Thanks to nlohmann for their [json](https://github.com/nlohmann/json) library used for project files
- Thanks to aquynh for [capstone](https://github.com/aquynh/capstone) which is the base of the disassembly window
### Contributors
- [iTrooz](https://github.com/iTrooz) for getting ImHex onto the Web as well as hundreds of contributions in every part of the project
- [jumanji144](https://github.com/jumanji144) for huge contributions to the Pattern Language and ImHex's infrastructure
- [Mary](https://github.com/marysaka) for her immense help porting ImHex to MacOS and help during development
- [Roblabla](https://github.com/Roblabla) for adding MSI Installer support to ImHex
- [Mailaender](https://github.com/Mailaender) for getting ImHex onto Flathub
- Everybody else who has reported issues on Discord or GitHub that I had great conversations with :)
### Dependencies
- Thanks a lot to ocornut for their amazing [Dear ImGui](https://github.com/ocornut/imgui) which is used for building the entire interface
- Thanks to epezent for [ImPlot](https://github.com/epezent/implot) used to plot data in various places
- Thanks to Nelarius for [ImNodes](https://github.com/Nelarius/imnodes) used as base for the data processor
- Thanks to BalazsJako for [ImGuiColorTextEdit](https://github.com/BalazsJako/ImGuiColorTextEdit) used for the pattern language syntax highlighting
- Thanks to nlohmann for their [json](https://github.com/nlohmann/json) library used for configuration files
- Thanks to vitaut for their [libfmt](https://github.com/fmtlib/fmt) library which makes formatting and logging so much better
- Thanks to btzy for [nativefiledialog-extended](https://github.com/btzy/nativefiledialog-extended) and their great support, used for handling file dialogs on all platforms
- Thanks to danyspin97 for [xdgpp](https://sr.ht/~danyspin97/xdgpp) used to handle folder paths on Linux
- Thanks to aquynh for [capstone](https://github.com/aquynh/capstone) which is the base of the disassembly window
- Thanks to rxi for [microtar](https://github.com/rxi/microtar) used for extracting downloaded store assets
- Thanks to VirusTotal for [Yara](https://github.com/VirusTotal/yara) used by the Yara plugin
- Thanks to Martinsos for [edlib](https://github.com/Martinsos/edlib) used for sequence searching in the diffing view
- Thanks to ron4fun for [HashLibPlus](https://github.com/ron4fun/HashLibPlus) which implements every hashing algorithm under the sun
- Thanks to mackron for [miniaudio](https://github.com/mackron/miniaudio) used to play audio files
- Thanks to all other groups and organizations whose libraries are used in ImHex
### License
The biggest part of ImHex is under the GPLv2-only license.
Notable exceptions to this are the following parts which are under the LGPLv2.1 license:
- **/lib/libimhex**: The library that allows Plugins to interact with ImHex.
- **/plugins/ui**: The UI plugin library that contains some common UI elements that can be used by other plugins
The reason for this is to allow for proprietary plugins to be developed for ImHex.
Supported is generally only the latest version that can be downloaded from the Release tab. If you have issues, you might get instructed to use the Nightly release version instead.
If you built ImHex yourself and experience issues that are not present in the version built by GitHub, you're on your own.
## Reporting a Vulnerability
Any critical vulnerabilities can be reported through Discord (@werwolv).
message(FATAL_ERROR"External dependency ${EXTERNAL_DIR} is empty!\nMake sure to correctly clone ImHex using the --recurse-submodules git option or initialize the submodules manually.")
# This is the official CMake template for making your own ImHex plugins
# To use this template, copy the file into its own directory and modify it to your needs
# For the most part, this is a regular CMake project with some extra functions provided by the ImHex SDK
#
# [NOTE FOR NON-C++ PLUGINS]
# The template is laid out for a C++ plugin, however you can write your plugin in any language you want
# and just make the plugin statically link against your code. The only thing that's required is a .cpp file with
# the IMHEX_PLUGIN_SETUP() macro used in it. This macro is used to setup the plugin and register it with ImHex
#
# [CMAKE FUNCTIONS]
# add_imhex_plugin(): Registers a new plugin
# NAME: The name of the plugin
# IMHEX_VERSION: The ImHex version this plugin is compatible with. If unset, the plugin will be loaded on all versions (this may not work though)
# SOURCES: Source files of the plugin
# INCLUDES: Include directories of the plugin
# LIBRARIES: Libraries to link against
# FEATURES: Optional features that can be enabled or disabled
# LIBRARY_PLUGIN: If set, turns this plugin into a library plugin. Library plugins can be linked against by other plugins
#
# add_romfs_resource(filePath romfsPath): Adds a file to the romfs of the plugin
# The RomFS is a virtual filesystem whose files can be accessed by the plugin using the functions in the `romfs::` namespace
# This function is used to add a single file to the romfs. You can however also simply create a `romfs` directory in your plugin directory and place your files and folders in there
# filePath: The path to the file on the disk
# romfsPath: The path to the file in the romfs
#
# enable_plugin_feature(feature): Enables a plugin feature
# Features are optional parts of the plugin that may or may not be available depending on build settings
# When a feature is enabled, `IMHEX_FEATURE_ENABLED(feature)` will be defined to true. Otherwise, it will be defined to false
# Use the `IMHEX_PLUGIN_FEATURES` macro in the main plugin file to define names to each feature and have them be listed in the plugin list
# feature: The name of the feature to enable
cmake_minimum_required(VERSION3.20)
project(ImHexPlugin)
# Include the ImHex SDK
# For this to work, you need to set the IMHEX_SDK_PATH environment variable to the path of the ImHex SDK
#
# On Windows, the SDK is next to the ImHex executable
# On Linux, the SDK is usually in /usr/share/imhex/sdk but this may vary depending on your distribution
# On MacOS, the SDK is located inside of the ImHex.app bundle under ImHex.app/Contents/Resources/sdk
if(NOTEXISTS$ENV{IMHEX_SDK_PATH})
message(FATAL_ERROR"The IMHEX_SDK_PATH environment variable is not set")
endif()
add_subdirectory($ENV{IMHEX_SDK_PATH}ImHexSDK)
# Register the plugin
# This will configure everything you need to make your plugin work
# Modify the arguments to your needs. Right now it defines a plugin called `example_plugin`
# with a single source file called `example_plugin.cpp` in the `source` directory
# By default you have access to the libimhex library to interact with ImHex
# as well as libwolv, libromfs, libfmt and ImGui, but you can link against any libraries you want
For a TLDR of commands see [How to build](#How-to-build)
# Introduction
The original CI we used (vanilla Github Actions) was great for specifying what steps to execute to build packages. It could even do some custom steps with reusable actions. But it had problem: no local reproducibility. This meant that:
- We couldn't test code properly locally, we were dependent on GitHub to do it
- If something was wrong and we had to debug the build script, it was *long and painful* because we had to wait for Github runners to finish builds, and couldn't quickly iterate
To solve this, we are now trying to move the CI build script to docker containers (so using Dockerfiles)
# How to build
Commands are available in the [CI](../../.github/workflows/build.yml) and you should prefer copying them from there.
But here is a general command that should work for every build we have:
Docker-based builds work with a Dockerfile. You run the Dockerfile, it builds the package.
We are using a base environment (often given to us by dockerhub) (e.g. ubuntu:22.04) which is really just a root filesystem, and we then run shell commands in that env, just like a shell script
Docker-based builds have two kind of caches used:
- layer cache, which mean that if a layer (instruction) hasn't been changed, and previous layers haven't changed, it will not be run again
- a `COPY` layer will be invalidated if one of the file copied has changed
- mount cache, which are per-instructions mounts that will be cached and restored in the next run. Mounts on different folders will not collide
Docker cache tends to grow very quickly when constantly making changes in the Dockerfile and rebuilding (a.k.a debugging what's going on), you can clear it with something like `docker system prune -a`
In the command saw earlier:
-`.` is the base folder that the Dockerfile will be allowed to see
-`-f <path>` is to specify the Dockerfile path
-`--progress plain` is to allow you to see the output of instructions
-`--build-arg <key>=<value>` is to allow to specify arguments to the build (like -DKEY=VALUE in CMake)
-`--build-context key=<folder>` is to specify folders other than the base folder that the Dockerfile is allowed to see
-`--output <path>` is the path to write the output package to. If not specified, Docker will create an image as the output (probably not what you want)
-`--target <target>` specifies which docker target to build
On Windows, ImHex is built through [msys2 / mingw](https://www.msys2.org/)'s gcc.
1. Download and install msys2 from their [website](https://www.msys2.org/).
2. Open the `MSYS2 MinGW x64` shell
3. Clone the repo using `git clone https://github.com/WerWolv/ImHex --recurse-submodules`
4. Install all the dependencies using `./ImHex/dist/get_deps_msys2.sh`
5. Build ImHex itself using the following commands:
```sh
cd ImHex
mkdir build
cd build
cmake -G "Ninja"\
-DCMAKE_BUILD_TYPE=Release \
-DCMAKE_INSTALL_PREFIX="./install"\
-DIMHEX_USE_DEFAULT_BUILD_SETTINGS=ON \
..
ninja install
```
ImHex will look for any extra resources either in various folders directly next to the executable or in `%localappdata%/imhex`
For low RAM-usage system, you can use `mingw32-make -j N install` instead, to reduce RAM usage at compile time. Where `N` is amount of jobs you are willling to run at once. Roughly ~1 GB of RAM usage per job.
<summary>A Hex Editor for Reverse Engineers, Programmers and people who value their retinas when working at 3 AM</summary>
<description>
<p>ImHex is a feature-rich Hex Editor aimed towards Reverse Engineers working with foreign data formats, malware, executables and raw memory.
Besides all the features a common Hex Editor has, ImHex also features a custom scripting language used to declare and dissect data structures, support for running YARA rules, a node-based graphical data pre-processor and support for various data sources such as files, raw disks or GDB Servers.</p>
# Create a file dedicated to store wasm size, because I know no way to get the wasm content length if the web server uses compression
# See https://stackoverflow.com/questions/41701849/cannot-modify-accept-encoding-with-fetch https://github.com/AnthumChris/fetch-progress-indicators/issues/13
RUN du -b /build/imhex.wasm | cut -f1 > imhex.wasm.size
// NOTE: This file creates a service worker that cross-origin-isolates the page (read more here: https://web.dev/coop-coep/) which allows us to use wasm threads.
// Normally you would set the COOP and COEP headers on the server to do this, but Github Pages doesn't allow this, so this is a hack to do that.
/* Edited version of: coi-serviceworker v0.1.6 - Guido Zuidhof, licensed under MIT */
// From here: https://github.com/gzuidhof/coi-serviceworker
if(request.mode==="no-cors"){// We need to set `credentials` to "omit" for no-cors requests, per this comment: https://bugs.chromium.org/p/chromium/issues/detail?id=1309901#c7
e.respondWith(handleFetch(e.request));// respondWith must be executed synchonously (but can be passed a Promise)
});
}else{
(asyncfunction(){
if(window.crossOriginIsolated!==false)return;
if(!('serviceWorker'innavigator)){
alert("Your browser doesn't support service workers.\nIf you're using Firefox, you need to not be in a private window.")
}
letregistration=awaitnavigator.serviceWorker.register(window.document.currentScript.src).catch(e=>console.error("COOP/COEP Service Worker failed to register:",e));
if(registration){
console.log("COOP/COEP Service Worker registered",registration.scope);
registration.addEventListener("updatefound",()=>{
console.log("Reloading page to make use of updated COOP/COEP Service Worker.");
window.location.reload();
});
// If the registration is active, but it's not controlling the page
<pathstyle="opacity:0.2"d="M 5.3808594,5 C 4.6158118,5 4,5.6158118 4,6.3808594 V 13.619141 C 4,14.384188 4.6158118,15 5.3808594,15 H 31.619141 C 32.384188,15 33,14.384188 33,13.619141 V 6.3808594 C 33,5.6158118 32.384188,5 31.619141,5 Z M 40.400391,5 C 39.624791,5 39,5.6247906 39,6.4003906 V 13.599609 C 39,14.375209 39.624791,15 40.400391,15 H 58.599609 C 59.375209,15 60,14.375209 60,13.599609 V 6.4003906 C 60,5.6247906 59.375209,5 58.599609,5 Z M 5.3808594,50 C 4.6158118,50 4,50.615812 4,51.380859 v 7.238282 C 4,59.384188 4.6158118,60 5.3808594,60 H 31.619141 C 32.384188,60 33,59.384188 33,58.619141 V 51.380859 C 33,50.615812 32.384188,50 31.619141,50 Z"/>
<pathstyle="fill:#ffffff;opacity:0.1"d="M 5.3808594 4 C 4.6158118 4 4 4.6158118 4 5.3808594 L 4 6.3808594 C 4 5.6158118 4.6158118 5 5.3808594 5 L 31.619141 5 C 32.384188 5 33 5.6158118 33 6.3808594 L 33 5.3808594 C 33 4.6158118 32.384188 4 31.619141 4 L 5.3808594 4 z M 40.400391 4 C 39.624791 4 39 4.6247906 39 5.4003906 L 39 6.4003906 C 39 5.6247906 39.624791 5 40.400391 5 L 58.599609 5 C 59.375209 5 60 5.6247906 60 6.4003906 L 60 5.4003906 C 60 4.6247906 59.375209 4 58.599609 4 L 40.400391 4 z M 5.3808594 49 C 4.6158118 49 4 49.615812 4 50.380859 L 4 51.380859 C 4 50.615812 4.6158118 50 5.3808594 50 L 31.619141 50 C 32.384188 50 33 50.615812 33 51.380859 L 33 50.380859 C 33 49.615812 32.384188 49 31.619141 49 L 5.3808594 49 z"/>
<metaname="description"content="Free and extremely powerful Online Hex Editor for your Web Browser. ImHex is a free and open source Hex Editor for Reverse Engineers and Developers and Data Analysts.">
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
