The previous method only worked locally inside the repo, but not once the package is published to npm.
Once published, built-in themes are flattened instead of nested.
- This component was not published to the rpm registry anyway
- We depended on the repo’s master branch which can break things for everyone
- Its repo was not getting much attention which meant
* no tests, no coverage report
* no linting
* no package linting (e.g. had 2 dependencies that weren’t actually used)
It will be simpler to manage this way
If a theme has a build process, it will likely publish a subfolder (e.g. “dist”) to npm.
It can simply set “thumbsup.themeRoot” in its package.json so thumbsup looks in the right place.
fix(index): Continue on errors during find
This ensures the gallery is created even when some files are
inaccessible (for example due to invalid names or file permissions).
Requires running the test in Docker as non-root to allow file access test
Thumbsup used to stop at the first problem processing a file.
This was a problem on large galleries where you’d need to run it again and again, fixing files as you went.
This change:
- skips problematic files and shows a summary at the end
- logs all warnings/errors to <thumbsup.log> when running the default output
Also refactor and cleanup of the logging logic.
GraphicsMagick can’t process animated GIFs well unless all frames are coalesced, which creates very large files.
This means there is a new dependency on Gifsicle which is very good at working with animated GIFs.
- Thumbnails are still processed through GraphicsMagick (more options, e.g. centre-crop)
- Large previews of animated GIFs go through Gifsicle
The startup checks are updated to check for the Gifsicle binary.
Fixes#96.
The index is still updated, and the media folder is still scanned to log how many tasks are required.
This is useful to know what thumbsup is about to do, without actually running the expensive tasks.
- %path expands to the path of the photo/video
- %keywords expands to the IPTC keywords of the photo
- {format} expands to the photo's EXIF date, e.g. {YYYY MM}
This will help understand usage patterns to know what to focus on, e.g.
- are many people using thumbsup on Windows?
- are there many galleries with > 10,000 photos?
The current code doesn't create an output structure for them, so we don't create thumbnails.
This is good since the thumbnail generation would likely fail.
However we still try to render thumbnails in the themes.
The themes could be smart enough to skip invalid files, but it's easier to ignore them from the start.
1. Move from a JSON index to a SQLite database.
- This allows the indexing to be interrupted & resumed
- Updating the index consumes less RAM than loading / saving an entire JSON object
- Loading the index consumes less RAM since it can be streamed, only exacting the properties we need every time (instead of loading all EXIF data in memory, only to discard most of it later)
- These make a big difference when processing 10,000+ photos
2. Switch from <glob> to a manual <readdir>
- Glob would take several hundred or GB of RAM when asked to find several thousand files
- Manual approach with <micromatch> library does the same thing in a fraction of the time / memory usage
3. Exiftool optimisations
- Run 1 exiftool process per CPU, still in batch mode (divide all files to be read into 1 bucket per CPU)
- Stream the exiftool output instead of buffering it in memory
Before changing it back, need to list the rules for extension changes. For example:
- should GIF thumbnail should be JPG, to avoid animations on the album page?
- what about transparent GIFs, will they look weird in JPG?
- maybe GIFs should stay as GIFs, but kept to a single frame only for thumbnails
- same thing for pngs, which might be better kept as PNG for transparency
- all other non-browser-friendy formats should become JPG
These rules will be a lot easier to implement when the new input data structure is in place