- Fix issue with authenticating API credentials via Requests. (#20)
- Disable multiprocessing for now. (#21)
- Confirm Python 3.8 support.
- Improve command-line message content and formatting.
- Confirm Python 3.6 and 3.7 support.
- Expand number of unsupported filetypes to exclude from import list.
- Improve CSV output file name formatting.
- Documentation edits and updates.
- Delete SQLite database if, upon program exit, it contains no records. (#14)
- Add some color, improve readability in console output. (#18)
- Add colorama dependency for printing ANSI escape code colors in Windows.
- Provide
User-Agent
andFrom
HTTP header fields. - Correct issue where
force_ocr
andsecure
parameters were not being set default tofalse
in the Ruby way. (#17) - Handle 50X errors from the API (which do not return JSON). (#15, thank you, Tom Meagher!)
- Add
update_processed_files
method to get page, file hash and other data that's not available upon upload. - Add
pages
,file_hash
,id
andtitle
to items tracked in database. - Create database upon
DocumentCloudUploader
initialization. - Get rid of the file extension in the document title. (#13)
- Bug fix: Properly test for presence of data and title keyword arguments.
- Remove extra line space on csv dump in Windows.
- Report number of files to be uploaded before starting.
- Better reporting of upload progress and results.
- More comprehensive filetype exclusion list.
- Record pdf and text URLs in database.
- Uploads that return status codes other than 200 are handled.
dump_to_csv
outputs contents of SQLite database. (#2)- Add
force_ocr
parameter to upload options. - Removed multiprocessing support for Windows for now.
- Report when upload file directory does not exist.
- Better testing for prohibited file types.
- Only create database after file path verified.
- Packaged for release to PyPi.
- Add multiprocessing. (#1)
- Exclude files of 400MB or larger from upload. (#3)
- Add initial tests.
- Scaffolding for documentation.
- Pre-alpha prototype