Output & Archives
Last updated
Find downloaded files, choose an archive format, and understand metadata, deduplication, and resume.
On this page
Each series has its own output folder, with chapter archives and shared metadata. Use the layout below to find files, then choose a container and compression mode.
Directory layout
<output-dir>/ # --output (default: ~/Downloads/comic-dl)
<Series Title>/ # one folder per series
ComicInfo.xml # series-level metadata
cover.jpg # series cover (when available)
<Chapter Title>.cbz # one archive per chapter
.comic-dl/
trash/ # removed series, purged after 7 days
.nomedia # Android gallery exclusion markerArchive formats
Choose the container with --format or [archive] format in config:
| Format | Extension | Container | Notes |
|---|---|---|---|
cbz |
.cbz |
ZIP | Default. Most widely supported by comic readers. |
zip |
.zip |
ZIP | Plain zip. Still embeds ComicInfo.xml. |
cbt |
.cbt |
TAR | Never compressed. tar -tf to list, tar -xf to extract. |
Switching formats never produces a duplicate chapter — existing downloads of any format count as “already downloaded.”
Compression
Control compression with --compress or [archive] compression:
| Mode | Effect |
|---|---|
stored |
No compression (default). Fastest; pages are written as-is. |
deflate |
Zlib deflate at default level. ZIP formats only. |
deflate:0-9 |
Deflate at a specific level (0=fastest, 9=smallest). |
Comic pages are already-compressed raster images, so deflate rarely shrinks
the archive. The default is deliberately stored; compression is opt-in.
.cbt (tar) archives are never compressed.
Inside the archive
- Page images — numbered sequentially:
Page_0001.jpg,Page_0002.png, … - ComicInfo.xml — chapter-level metadata: series, chapter title, number, volume, page count, source URL.
Shared series metadata (description, creators, genres, rating) is included in
each archive’s ComicInfo.xml when available, as well as in the folder-level
ComicInfo.xml beside the archives.
See Metadata for the full field mapping.
Duplicate detection
SHA-256 dedup runs inside each archive. When multiple pages share the same byte size, only the first page with a given hash is kept. The rest are skipped. This is deterministic — the first page always wins.
Atomic writes
Archives are written to a .tmp file, verified (testzip() for ZIP, full
member read for TAR), then atomically renamed. A partial archive is never
left behind on disk.
Partial downloads and resume
If some pages fail (e.g. a temporary 404), the chapter is saved as a partial
CBZ. A <chapter>.cbz.partial marker is written next to it and the run exits
non-zero. Partial chapters are reported separately from finished ones — the
end-of-run summary shows a Download incomplete/Download completed with errors verdict and a (N partial) count rather than folding them into
failed.
The next run treats that chapter as not downloaded and retries just the
missing pages, overwriting the partial file. Re-running without --force
repairs incomplete chapters.
Text-only galleries
Galleries with no images are saved as <Chapter Title>.md (Markdown) with a
<!-- source: URL --> marker as the first line.
FSIComics folder grouping
FSIComics chapter URLs containing a -chapter- marker in the slug are
grouped under a Series Title – Artist folder. Single-chapter downloads keep
the plain Series Title folder.
Removed series
Removed series move to <output-root>/.comic-dl/trash/ and are purged after 7
days. See Remove a series and
Restore a series for the commands.
Temporary files
Scratch space is created under your platform temp directory during each run and cleaned up automatically on completion or interruption:
| Platform | Location |
|---|---|
| Linux/macOS | /tmp/comic-dl/ |
| Windows | %TEMP%\comic-dl\ |