📰 News

v5.1.2

• Features

  • The end-of-run Metron summary counts connection failures and responses that arrived without rate-limit headers.

• Performance

  • A Metron search that misses now costs 2 API requests instead of up to 6.

v5.1.1

• Features
  • A run that stops itself says why. OnlineSession.abort_reason carries the reason a lookup aborted the batch — most usefully a spent daily API quota, which stops a run nobody asked to stop. A caller's own cancel() leaves it unset, so an embedding application can tell a pause it initiated from one the server forced.
  • A search the daily-quota reserve stopped now reports Skipped with reason="quota_reserved" instead of passing for a source that looked and found nothing. Those two call for opposite handling: a comic nobody searched for should be tried again tomorrow, and one that genuinely missed should not. The matcher's own reason is now the exported constant SKIP_MATCHER_DECLINED alongside SKIP_QUOTA_RESERVED.

v5.1.0

• Fixes

  • A comic with no issue number no longer mis-tags itself with the first issue of a series matched earlier in the same run.
  • Metron tagging no longer sends requests the server is going to refuse. Every request now passes through a process-wide rate gate that models Metron's own sliding window, so a -j N batch stays inside the per-minute limit instead of discovering it by being rejected. This matters more than it sounds: Metron counts a rejected request against the daily quota exactly like a successful one, so the old behavior spent budget on nothing. Reported as #207.
  • A rate-limit rejection now releases one worker when one slot frees, rather than every worker at once. Retry-After says when a single request becomes possible, not when the window refills; waiting it out and then sending the whole queue earned one success and a fresh burst of rejections.
  • online.tuning.retry_budget is wired up. It was parsed, defaulted and then never passed to the retry decorator, so a config that set it kept getting the built-in 5 attempts; it now gets the number it asked for.
  • An aborted lookup is no longer retried. Nothing classified OnlineLookupAbortedError, so it fell through to the retriable default and was replayed on the rate-limit schedule.

• Features

  • Comicbox warns when Metron's daily quota drops below 10%, stops starting new searches near the floor so the rest of the budget finishes comics that already matched, and stops cleanly at zero. Re-running is cheap — the response cache replays what was already fetched.
  • The end-of-run summary reports API cost: requests sent per endpoint, rejections, seconds spent pacing, and the remaining per-minute and daily budgets.
  • online.tuning.per_source.metron.rate_limit.per_minute works again, as a ceiling on the server-reported limit. Set it to share one token between several processes. Note that a config which already sets it was being warned it was ignored, and will now actually be held to it. per_day is still ignored and still says so.
  • --jobs is no longer capped at 20 for Metron runs. The cap bounded workers, which is not what Metron throttles; the gate bounds requests.
  • The outgoing User-Agent identifies the entry point, e.g. comicbox/5.1.0 (cli; jobs=8). OnlineSession(client_name=...) names an embedding application.
  • metron_requests_for_batch(comics, series) prices a batch the way one actually runs — each series resolved once, the rest of its comics at the warm price — instead of charging every comic a cold search. Pacing math now prefers the burst limit Metron reported over the documented default.

• Performance

  • A -j N batch resolves each series once instead of once per worker. Workers that want the same series wait for whoever is resolving it and then take the warm path, and the dispatcher hands the pool one file per series first, so its first N tasks are N different series rather than N copies of the same cold search.
  • Metron batches holding many comics from one series fetch that series' issue list once instead of once per comic, when Metron's reported issue count says that is actually cheaper. A 100-issue single-series run drops from about 200 requests to about 102.
  • Comic Vine tagging spends half the API budget it used to. simyan 4.1.0 stops paginating past a short page, so each candidate volume's issue lookup costs one request against Comic Vine's 200-per-hour issues pool instead of two. That pool paces a Comic Vine run, so runs are faster as well as cheaper, and estimate_run() now projects the request count rather than a floor half its size.

• Dev

  • Require simyan >= 4.1.0.

v5.0.0

• Breaking Changes

  • Comicbox schema v3.0. Version 2.0 documents no longer load.
  • Credit roles and age ratings use MetronInfo's names and scale, so finer ComicInfo ratings like Everyone 10+ become Everyone. The primary credit flag is now per role, not per person.
  • manga no longer carries reading direction. See the new reading_direction and manga_volume fields.
  • Web links moved out of identifiers into a new urls list, and identifier_primary_source became primary_id_source.
  • ComicInfo's GTIN holds a barcode again instead of a comicbox urn, which is what Kavita, Komga and Mylar expect. Files written the old way read.
  • A stated title is kept as written and stories are only read out of it when the comic lists none. A series' other names live in series.alternative_names.
  • Removed critical_rating and alternate_images, which mapped to no format.
  • Removed --replace; use --merge-mode update. The write API takes merge_mode instead of mode, and WriteMode is now MergeMode.
  • OnlineSession takes match and prompts instead of mode and unattended, matching --match and --prompts. set_mode and set_unattended are set_match and set_prompts, the prompt objects carry match and prompts, and the set_unattended selector action is set_prompts, which names the policy it sets and so can turn prompts back on.
  • Environment variables nest with __, and every config key can be set that way: COMICBOX_ONLINE__AUTH__METRON__KEY. Old flat names warn.
  • estimate_run() and requests_per_comic() take an effort instead of a match mode, which never changed what a search costs.

• Features

  • New --merge-mode chooses how supplied metadata merges into a comic's existing tags: additive (the default), replace or update.
  • Web urls in the Notes field are read into urls.
  • Online API: new SourceStarted event, Effort is exported alongside MatchMode, and rate_limit_status() now covers Comic Vine as well as Metron.
  • OnlineSession accepts a config, the settings it layers its tagging preferences over. An embedder that already holds configured settings — naming its own cache directory or effort — hands them over instead of exporting environment variables for the settings loader to find.

• Fixes

  • Two silent data losses: an XML tag carrying an attribute, like <Series lang="en">, took the file's whole metadata down with it, and MetronInfo tag-level ids were discarded in files that name no primary id source.
  • Writing metadata no longer rewrites a comic's pages, so tagging is faster and an interrupted write can't damage a page. One archive named twice in a batch is no longer repacked by two threads at once and destroyed.
  • Online tagging reports a match only when metadata was really applied, and a lone mediocre match no longer auto-writes without a prompt. Prompts hold up under batch and parallel runs.
  • --effort balanced is honored; a large unattended run no longer quietly downgrades it to minimal for Comic Vine.
  • -c/--config loads the file it names instead of being ignored, and an unknown --online source is an error rather than a silent widening: a typo like --online metrn queried every configured database.
  • Reading metadata no longer depends on write settings, so --online --write … --rename no longer names the file from stale data.
  • A comic that can't be read no longer ends the batch, and comicbox exits non-zero whenever any file failed.
  • MetronInfo alternative name ids link to their series, and reprint ids to the issue they reprint, instead of to pages that don't exist.
  • The no-TTY hint names --prompts never instead of --unattended, which never existed.
  • Many smaller repairs to credit roles, tag coverage, identifiers, urls and odd data that used to cost a whole comic. Metadata comicbox skips is now named in a warning instead of dropped silently.

• Performance

  • A 200 page, 117 MiB CB7 converts in 6.2s instead of 232s, with 318 MiB peak RSS instead of 1.9 GiB.
  • Online tagging batches by series, downloads covers in parallel and bounds Comic Vine calls per comic (--effort thorough restores the unbounded search). Rate-limit waits are interruptible again.
  • Comicbox starts about a third faster and building a Comicbox is about forty times cheaper. Books with many reprints read up to fifty times faster.

• Dev

  • Require simyan >= 4.0.0.

v4.8.6

• Features

  • Comicbox.predict_filename() returns the scheme name rename_file() would rename to, without renaming. rename_file() now uses it, so a caller can show or check the destination first.

• Fixes

  • Renaming keeps the archive's own file extension. The name ends in ext, a metadata field, so a config that skips it fell back to "cbz" and a stale value embedded by another tagger overrode the real one — either way renaming a PDF or CBT to a name claiming to be a zip.
  • Converting no longer writes the pre-conversion ext into the new archive's metadata, which is what left those stale values behind.
  • Archives differing only by suffix no longer destroy each other when converted in one bulk write. They all convert to the same .cbz, and the loser could replace the winner's finished file while delete_orig unlinked both originals; it now fails cleanly instead.

v4.8.5

• Features
  • write_metadata and bulk_write results report final_path: the archive's on-disk path after the write. It differs from the submitted path when the write repacked an unwritable archive (CBR/CBT/CB7) into a CBZ, so callers tracking files on disk can follow the conversion.

v4.8.4

• Features
  • Short options for more common flags. See --help.

v4.8.3

• Fixes
  • CBR files whose RAR3 extended timestamps carry a sub-second remainder of a second or more no longer fail to open with ValueError: microsecond must be in 0..999999. rarfile <= 4.5 overflows parsing these, which some third-party packers write; comicbox now carries the excess into whole seconds. The patch deactivates itself if a future rarfile fixes the overflow.
  • Content based archive detection tries formats in observed frequency order (zip, rar, pdf, 7z, tar) when the file extension hint misses, instead of trying the rarest formats first. Every detector in that order now requires leading magic — zip included — so a pdf or 7z with zip data appended can no longer be claimed by the zip tail scan; zips with prepended data (self-extractors) are still detected by a terminal loose pass.

v4.8.2

• Features
  • Metron tagging populates language from the issue's series.
  • Metron prices in EUR and ITL are attributed to Italy. EUR is deliberate: metron.cloud's euro prices currently all come from its Italian comics catalog.
• Dev
  • Require mokkari >= 4.5.0.

v4.8.1

• Dev
  • Require simyan >= 3.1.0 and use its new search_volumes() API.

v4.8.0

• Features
  • OnlineSession accepts pinned per-source issue ids. A source with an id fetches it directly while the other enabled sources search, so one session can mix id retrieval and search and merge both results.

v4.7.1

• Features
  • --auth takes an API token directly, as --auth metron:TOKEN or --auth comicvine:KEY.
  • Metron username and password authentication is deprecated and now warns. It will be removed in a future release; generate a token on your metron.cloud account page.

v4.7.0

• Features
  • Metron accepts an API token instead of a username and password. Set it with --auth metron:key=TOKEN, the COMICBOX_METRON_KEY environment variable, or the online.auth.metron.key config key. The token wins if both are set.

v4.6.1

• Fixes
  • Hand-tagged identifier keys with a type prefix, like series:178012, now produce correct per-database URLs instead of embedding the prefix in an issue-shaped URL. Affected every source, notably ComicVine, Metron, Grand Comics Database, and League of Comic Geeks.
  • Hand-tagged ComicVine long codes like 4050-160294 are normalized to the bare id with the right URL.
  • source:type:key identifier strings in tags no longer drop the id number.
  • All URNs in a notes field are read as identifiers; previously only the first was.
  • Story arcs tagged online from Metron or ComicVine get web URLs.

v4.6.0

• Features
  • Metron and ComicVine online tagging record a series' alternative names (localized and variant titles) as alternate-series reprints, the same way MetronInfo.xml's Series/AlternativeNames are read. Metron alternative names arrive with mokkari 4.3.0; ComicVine's come from volume aliases.
  • Online matching recognizes ComicVine volume aliases. A comic filed under a localized or variant series title now matches its canonical volume, both in candidate scoring and in the pre-call volume filter.

v4.5.1

• Fixes
  • ComicInfo.xml's BlackAndWhite tag works in both directions. The Yes value was missing from its enum, so monochrome could never be read from a comic nor written to one, leaving the field impossible to set or clear. The schema's Unknown reads as no value.
  • Enum tags that accept booleans (BlackAndWhite, Manga) now handle False as well as True. Writing either previously raised and dropped the tag.

v4.5.0

• Features

  • The public write API can now clear fields: write_metadata() and BulkWriteItem accept delete_keys, glom key paths removed from the final metadata before writing (layered onto the config's general.delete_keys). Empty patch values are pruned on schema load, so previously an update-mode write could never clear an existing tag. The patch may be empty when delete_keys is non-empty (a pure-clear write).
  • Extracting a range of PDF pages as PDFs now writes one PDF of the whole range instead of a file per page.

• Fixes

  • Default PDF to CBZ conversion produced an archive of raw .ppm pixmaps that comic readers (including comicbox itself) do not recognize as pages, yielding a zero page book — and --delete-orig then removed the working original. Conversion now defaults to pixmap_jpeg whole-page renders, which also apply pdf display rotation.
  • --pdf-pages image extraction and conversion wrote rotated pages as stored — sideways or upside down for scans that rely on /Rotate (or a rotated content-stream placement) for display. Rotated image-dominant pages are now rendered to match the displayed orientation; unrotated pages still extract their original bytes untouched.
  • Converted pdf pages are now stored uncompressed in the zip like other images; the compression decision previously ran before pages gained their image suffix, so page images were pointlessly deflated.
  • The --pdf-pages help lists exactly the values it accepts. It advertised pdf where unsupported and omitted image_if_dominant and pixmap_jpeg.
  • A PDF with embedded metadata (e.g. a written ComicInfo.xml) no longer counts that file as an extra page, which inflated the page count and left a stray file when extracting page ranges.
  • Extracted PDF pages are named for their contents. Pages extracted as images were given a .pdf suffix.

v4.4.0

• Features
  • Support the MetronInfo.xml v1.1 schema: new AlternativeNumber and CommunityRating (AverageRating, RatingCount) tags.
  • New comicbox community_rating field (average_rating, rating_count). ComicInfo CommunityRating and ComicBookInfo rating now map to it instead of critical_rating, which remains but no longer maps to any format.
  • New comicbox alternative_issue field, parsed into name, number and suffix parts like issue. The Metron API alt_number now maps to it instead of the issue suffix.
  • Metron online tagging now fills community_rating from the API's average_rating and rating_count (mokkari 4.2.0).

v4.3.0

• Features
  • OnlineSession.rate_limit_status() is no longer a stub: Metron entries now report the account's live rate-limit budget (burst and daily sustained windows — limit, remaining, and reset epoch) that mokkari 4 tracks from X-RateLimit-* response headers. The daily limit varies by Metron OpenCollective donor tier, so this is the only place it's discoverable. Comic Vine still reports {}; simyan exposes no budget to read. Host applications (e.g. Codex) can poll this during a batch run to display live budget.

v4.2.0

• Config

  • Custom Metron rate limits (per_minute/per_day) in the config are no longer supported and will be ignored in favor of the limits Metron reports for your account — mokkari 4 reads them from API response headers, so higher OpenCollective donor tiers are picked up automatically. You will see a warning if you still have them set. These keys are deprecated for every source and will be removed in a future release.
  • --jobs is capped at 20 (Metron's burst limit) when Metron is an active credentialed source for the run; the log explains the cap when it engages.

• Fixes

  • Rate-limited API responses missing a Retry-After header now back off on the normal schedule instead of retrying immediately.
  • Ignored-config warnings (rate-limit overrides, --api-url for Metron) now print once per run instead of once per file.

v4.1.1

• Performance

  • Upgraded Simyan, the ComicVine API library, for more accurate wait-time estimates, fewer hangs, and automatic background cleanup of old cached data.

• Config

  • Custom ComicVine rate limits (per_second/per_hour) in the config are no longer supported and will be ignored in favor of the official limits. You will see a warning if you still have them set.

v4.1.0

• Features
  • Batch API callers can project the API-request count and duration of an online-tag run via comicbox.online_session.estimate_run(), so the request model lives with the search flow it describes.

v4.0.5

• Performance
  • Metron online tagging now identifies an issue in far fewer API requests — often one or two instead of dozens — so tagging is faster and more reliable. Courtesy of @bpepple. The --effort option no longer applies to Metron due to this improved API.

v4.0.4

• Fixes
  • Online tagging now searches using embedded metadata (e.g. series name) instead of the filename when both are present, so already-tagged comics match more accurately.
  • Renaming no longer crashes when metadata contains a path separator (e.g. a slash in the series or title); the separator becomes _ in the name.

v4.0.3

• Features

  • Add Sweet Shop kebab case filename format. Courtesy of @bpepple.

• Fixes

  • Online cover matching now works for PDFs; the cover page is rasterized before perceptual hashing instead of handing raw PDF bytes to the image hasher. Courtesy of @bpepple.
  • Security release for dependencies.

v4.0.2

• Fixes
  • CBR files whose RAR entries carry sub-microsecond timestamps no longer break the ProcessPoolExecutor worker pool. rarfile's nsdatetime is an unpicklable datetime subclass; its mtime is now coerced to a plain datetime so results survive transfer back from worker processes.

v4.0.1

• Fixes
  • A source's web URL no longer clobbers its authoritative <ID>/GTIN identifier key on read. Metron/Comic Vine ids now survive a write→read round trip in ComicInfo and MetronInfo, so already-tagged comics refresh via their stored id instead of falling through to a full online search.
  • Metron url parsing no longer captures a trailing slash into the id key (.../issue/123495/ -> 123495, not 123495/).
  • The AutoWritten event now fires for matches resolved by explicit id or stored id, so it reports a source for every auto-written comic.

v4.0.0 - Online Metadata Tagging

• 🚨 Breaking Changes
  • Removed the ComicTagger (comictagger.json) format.
  • Reorganized config files into nested groups (general, read, write, convert, online, …). Flat v3 config files must be migrated.
  • Dry-run is now -n (was -y).
  • PDFs no longer decode comicbox metadata hidden in the PDF keywords field; keywords now reads as plain tags. Metadata embedded as files inside a PDF still reads normally.
  • Library API: Comicbox(...) no longer accepts a logger= argument and never reconfigures the host application's logging. Host apps keep their own loguru sinks; the CLI configures its own.
  • Library API: operational errors derive from comicbox.exceptions.ComicboxError (ArchiveError, MetadataError, ExportError, …) instead of raising bare ValueError. Consumers can except ComicboxError.
• Fixes
  • Normalize ComicBookInfo rating to the canonical 0-5 scale.
  • Map Metron price currencies to the correct country.
  • Harden credentials against accidental leaks into logs.
  • write_metadata() accepts the root-wrapped dict to_dict() returns; passing it used to silently write nothing while reporting success.
  • Metadata added with add_metadata() after a read is no longer ignored.
  • to_dict() computes pages and page count correctly when one instance dumps multiple formats.
  • Exported metadata files no longer end with a trailing blank line.
• Features
  • Online metadata tagging from Metron and ComicVine with --online. Comicbox searches by the comic's series, issue, and year, ranks the candidates, breaks close calls with cover-image matching, and writes the best match.
    • --match {ask | careful | auto | eager} sets how confidently a match is written without asking; --prompts never turns prompts into skips for cron and batch runs.
    • --id <db>:<id> tags by an exact upstream id; --series-id <db>:<id> constrains a search to a known series.
    • Credentials resolve from --auth, COMICBOX_* environment variables, the config file, or the system keyring.
    • Online responses are cached on disk (--cache, --cache-ttl); --effort trades API calls for matching accuracy.
  • Process many files in parallel with -j N, including batch online tagging.
  • New public write API: comicbox.write.write_metadata and bulk_write.
• Performance
  • Faster archive reads: the 7z and RAR backends load lazily, the page list is derived from cached archive state, and that state is released on close().

v3.0.3

• Fix small crashes with metron credits and comicbox with no path

v3.0.2

• Use new comicfn2dict. Parses more comic filename variations.

v3.0.1

• Require new comicbox-pdffile that doesn't corrupt PDF pages. Includes image-dominant page detection (PDFFile.classify_page, PDFFile.read_image_if_dominant, PDFFile.read_full_pixmap_jpeg) used by browser readers to serve scanned-comic PDF pages as plain <img>.

v3.0.0 - Config Dataclass & Parallel Reads

• Breaking Changes
  • get_config() now returns a ComicboxSettings dataclass, not a Confuse AttrDict. Comicbox constructor now accepts this dataclass instead of an AttrDict
• Fixes
  • Stop emitting metron.cloud/{genre,location,reprint,role,story,tag}/... URLs for Metron identifiers — those paths 404 because Metron has no public web pages for those types (only API endpoints). The numeric Metron ID is still preserved on the identifier.
  • Security against suspicious archive paths when extracting pages and metadata to the filesystem.
• Performance
  • Reducing startup time for new instances of comicbox.
  • General performance improvements for reading metadata from many files.
  • Special multiprocessing and async methods comicbox.process.iter_process_files() and comicbox.process.aread_metadata() for reading large batches of files at once.
  • Comicbox.get_cover_page(skip_metadata=True) skips metadata parsing for callers that just need the first archive image as a thumbnail. Removes per-call schema instantiation and Union resolution overhead.
• Features
  • Add Age Rating conversion function comicbox.enums.maps.to_metron_age_rating(value: str | Enum) -> MetronAgeRatingEnum | None

v2.2.3

• Update marshmallow & rich

v2.2.2

• Update deepdiff library

v2.2.1

• Fix reading PDF metadata breaking on datetimes.
• PDF datetime parsing failures are warned about instead of raising an exception and abandoning parsing.

v2.2.0

• Add --pdf-page-format option to choose how pdf pages are extracted. Replaces to_pixmap in the API.
• --cbz now works on PDF files. Can create original image CBZs with --pdf-page_format image for PDFs with single image pages.
• --validate validates imported and archive metadata against available schemas
• --import expands quoted "glob*" paths.
• Fix comicbox jsonschema.
• Many more type annotations. Include py.typed sentinel file.

v2.1.1

• Pin confuse config library at 2.1.0 due to typing issues with 2.2.0
• YAML formats now dump unquoted date and datetimes.

v2.1.0

• PDFs now write to pdf embedded files instead of overloading the metadata keywords field. Thanks for the idea from @bpepple
  • PDFS still read full metadata formats from the keywords field if they exist for backward compatibility.
• Read more flexible datetimes from PDFs

v2.0.6

• Python 3.14 support.

v2.0.5

• Fix a crash when dumping comicbox.json format with only one page.

v2.0.4

• The confuse library dependency doesn't support Python 3.14. So neither does comicbox :(

v2.0.3

• Fix ComicInfo.xml coercing Age Rating to wrong value.
• Normalize more original_format names.
• Update xmltodict. Removes much comicbox fix code.

v2.0.2

• Ensure mtimes from archives are timezone aware.
• More aliases for comicvine identifier source.
• Make archive comments that aren't ComicBookInfo JSON log as debug comments more often.

v2.0.1

• Resolve circular import if not installed with [pdf] option.

v2.0.0

🚨 BREAKING CHANGES 🚨

• Schema, API, config and CLI changes. See the 2.0 CHANGES document.

Features

File Formats

• CB7 archive read support. Comic archives in 7zip archives.

Schemas

• Support the MetronInfo.xml v1.0 Schema
• Add AniList, Kitsu, MangaDex, MangaUpdates, MyAnimeList identifier sources.
• ComicInfo.xml gains the Translator tag
• PDF modDate is now read and written.
• URNs as serialized identifiers in notes tag gain an optional tag type attribute in the nss: e.g. urn:comicvine:series:1234
• Comictagger schema supports series_aliases and title_aliases as reprints
• Parse PDF datetime format.
• Many enum fields now accept caseless and slightly fuzzy value lookups that are coerced to correct types for the specified output metadata format.
• For convenience, named or numbered types or collections in the comicbox schema may also be parse by their simple name instead of requiring a complex object.

Config

• --delete action becomes --delete-all-tags.
• --compute-pages is off by default. Turn on to recompute ComicInfo style Pages structures
• --no-compute-page-count prevents recomputing page_count.
• --delete_keys now excludes keys from loading entirely.
• Syntax highlighting used on output. Change or remove with --theme option.

Fixes

• ComicInfo.xml StoryArcs tag was not loaded.
• Fix proper lowercasing of serialized boolean values in xml attributes.
• ComicBookInfo issue tag becomes an integer.
• Fix ISBN & UPC url detection
• ComicInfo.xml orders tags properly per the xsd.
• Corrected schemaLocation tags for xml formats to be valid.
• ComicBookInfo.json rating, and tags tags fixed.
• Support ComicBookInfo.json primary credit tag.
• More accurate merging of different metadata sources.

v1.2.3

• Fix story arc parsing.

v1.2.2

• Fix Notes parsing for Comictagger beta Metron origin
• Simplify Identifier URL construction for Metron pk ids.

v1.2.1

• Fix ignoring MacOS resource forks in archives.
• Detect .jxl extension (JPEG XL) as a comic page.
• Dependency xml parsing fixes and better Python 3.13 support.

v1.2.0

• Add -R --replace_metadata option. Default behavior is to merge keys.
• Add -D --delete_keys option.
• Faster metadata writing. Replace files in zipfiles instead of rewriting the entire archive.

v1.1.10

• Fix parsing negative issue numbers in filenames.
• Log common non-ComicBookInfo archive comments with less alarm.

v1.1.9

• Deps security update

v1.1.8

• Update pycountry

v1.1.7

• Unknown urls give the path, query and fragment as the nss, not the domain

v1.1.6

• Ignore dotfiles and macOS resource forks when finding pages.

v1.1.5

• Fix export of CIX CoverArtist tag.
• Fix tagging Web and other unknown url tags. Accept any url for comicbox identifiers.

v1.1.4

• Make language and country code parsing more durable

v1.1.3

• Updated comicfn2dict fixes filename parsing bugs.

v1.1.2

• Fix crash copying directories between archives.
• Fix crash with leftover temp files.

v1.1.1

• Write ComicInfo.xml in TitleCase so Comictagger can find it.
• Write MetronInfo.xml & CoMet.xml in TitleCase for beauty.
• Fix duplicate reprints.

v1.1.0

• Fix --import option crash.
• Fix parsing tagger and updated_at from notes.
• PDFs read all metadata formats from the keywords field.
• PDFs write ComicInfo.xml to keywords field.
• ComicInfo.xml accepts spaces as web field separator.
• Accept numeric types for issues.
• Major improvements to filename parsing and diversity of filename schemas.

v1.0.0

• 🚨 This version contains large breaking changes 🚨
  • Some are detailed in The 1.0.0 CHANGES document
• Comicbox continues to primarily be an API for reading comic metadata but this version contains an enhanced CLI, and more powerful reading, writing, synthesis and exporting of metadata.
• Comicbox now sorts archive filenames case insensitively.
• Comicbox now writes identifiers to the notes field in urn format.
• You should probably read the code in comicbox.box for public facing apis.
• comicbfn2dict and comicbox-pdffile are now independent packages.

v0.10.2

• Sophisticated cli metadata parsing. See cli help and README.
  • config.metadata_cli holds the new string format.
• Writing xml and json metadata to files is now pretty printed.
• Fix some instances where falsey values were not written.
• Fix comicinfo.xml ComicPageInfo typing.

v0.10.1

• Change --metadata cli syntax to use key=[a,b,c] for arrays and key=a for simple values.

v0.10.0

• WARNING: Breaking API, CLI & Config changes.
• Write metadata from the command line.
• Optional PDF support. Install as comicbox[pdf]
• Python version now >=3.9
• Extract a range of pages from the cli.

v0.9.1

• Removed dependence on Python 3.11

v0.9.0

• StoryArc & StoryArcNumber for ComicInfo.xml exported as story_arcs dict. Supports Mylar multiple story arcs csv values format.

v0.8.0

• Add close_fd & check_unrar config options for API use.
• Add ComicArchive.check_unrar_executable() public method.
• Remove closefd constructor option.

v0.7.1

• Remove unrar.cffi support.
• Test for unrar executable for clearer errors.

v0.7.0

• Fix
  • Tags from ComicInfo.xml were not parsed
• Features
  • ComicInfo.xml StoryArcNumber, Review and GTIN now parsed.
  • ComicInfo.xml Pages attributes now exposed as snake case for python

v0.6.7

• Fix
  • Remove unrar-cffi dependency aciddentally left in during testing

v0.6.6

• Features
  • Print filetype cli option. get_archive_type() api method.
  • Use unrar.cffi if it's available.
• Dev
  • Use importlib instead of deprecated pkg_resources.

v0.6.5

• Feature
  • Demote parser errors to warnings.

v0.6.4

• Feature --delete-rar option is now --delete-orig
• Fix
  • Fix --delete-rar option sometimes deleting original cbzs

v0.6.3

• Fix

• Fix rename function renaming files to nothing.

• Fix --dry-run feature for rename.

v0.6.2

• Fix
  • Enable support for Deflate64.

v0.6.1

• Fix
  • Fix encoding/decoding crash by replacing uncodable utf-8 characters.

v0.6.0

• Features

  • Add --metadata cli action. Injects metadata from cli.
  • Warn when no actions performed.
  • == operator for metadata is deep and ignores key order.
  • Credits are now sorted.

• Fix

  • Log format conversions.
  • Don't add empty credits list metadata.

v0.5.5

• Fix
  • Fix dest-path cli argument.
  • Use defusedxml for XML parsing.
  • Fix recursion crash.
  • Log exception during recursion and proceed.

v0.5.4

• Fix
  • Remove uneccissary dependencies

v0.5.3

• Fix
  • Fix filename extension parsing
  • Renamed underscore cli options to use dash instead
  • Fixed crash when recompressing directories.

v0.5.2

• Features
  • ComicArchive class now has a context manager
  • Removed as_pil() methods for pages and covers

v0.5.1

• Features
  • Methods for getting covers and pages as PIL Images.
  • Lazy metadata and cover pulling. Removes metadata and cover options.
  • closefd option leaves archive open after method calls. Close manually with close().
  • .cbt Tarfile comic archive support.

v0.5.0

• Features

  • Issues numbers are now strings.
  • Separate read metadata option from print metadata action.
  • Added dry_run option.
  • Namespace config under "comicbox" map to allow inclusion in caller configs.
  • Allow modnames for local config sources, useful when comicbox is a library.

• Fixes

  • Trap errors reading user config files.
  • Fixed cover extraction to a directory.
  • Abort with message if pages asked to extracted to a filee.
  • Handle more filename patterns.

v0.4.1 - Yanked

v0.4.0

• Fixed some file name patterns
• Accept environment variables and config files to configure.
• CLI accepts more than one path as targets for action.
• CLI runs every action on the command line.
• API get_cover init variable now set in config as cover
• Optional metadata parsing with config.

v0.3.4

• Fixed combining CBI credits and with other format credits.
• More info in parsing warnings.
• More explicit raw tag titles.

v0.3.3

• add colored python logging

v0.3.2

• issue_count type change from decimal to int

v0.3.1

• Fix synthesizing CBI & CIX credits data.
• Extract decimals from strings for more liberal reading.

v0.3.0

• CBI changed critical_rating to decimal type from string (in CBI it's an int).
• CIX gained the community_rating attribute from CIX 2.0 spec.
• age_rating replaces maturity_rating to standardize more on CIX
• CoverArtist replaces Cover to standardize more on CIX
• Better CIX Manga parsing
• Better CIX BlackAndWhite parsing
• Added CIX AgeRating valid schema values, unused.
• LGPL 3.0

v0.2.2

• Update pycountry to be compatible with codex

v0.2.1

• Comicbox now raises a specific UnsupportedArchiveTypeError for bad archives

v0.2.0

• Option for getting the cover image at metadata load time as well.

v0.1.7

• Remove regex dependency in favor of re module

v0.1.6

• Fixed recursive mode trying to operate on directories
• Fix recompressing crash on cbr -> cbz conversion
• Fixed parsing for three new filename patterns
• Updated dependencies

v0.1.5

• Fix parsing and writing genre, story_arc, series_groups tags

v0.1.4

• Make metadata read more robust by catching individual tag exceptions and printing them to stdout
• Protect against null people names for credits
• More robust xml volume parsing

v0.1.3

• Fix cli crash with bad import

v0.1.2

• Fix relative imports
• Fix credit parsing errors
• Improve 1/2 Issue number parsing

v0.1.1

• Don't crash import on bad credits

v0.1.0

• initial release