docspell

Changelog • Documentation

This release comes with some smaller features:

• Feature: Remember-Me – another cookie is used to provide a remember-me functionality. The cookie is checked against some value in the database, so an admin can always make all remember-me cookies invalid. (#435)
• Feature: Link persons to organizations. In the address book, a person can now be associated to an organzition. The dropdowns show a little hint for which organization a person belongs to. Also suggestions for persons are restricted to those of the organization if that has been associated before. (#375)
• Feature: Allow to filter on source names. The search form can now search by a source name. The new field can be hidden via ui settings (it must be activated for exising users). (#390)
• Feature: Customize the title and subtitle of the item card in the overview. You can now define patterns for the title and subtitle of a card. (#429)
• Feature: Export your data. A bash script has been added that goes through your items and downloads them all to disk including their metadata!). This can be used to periodically backup the data in docspell.
• Improvement: The webui has been improved in that the search bar and search form are unified regarding the text search. The two fields in the form, allowing to search in names and fulltext, have been combined into a single field just as the search bar. (#497)

This is the last release for 2020. I had hoped to put more into this, but this time of the year is always a busy one ;-). I want to thank you for your support and interest in this project and I wish you all a joyful Christmas time!

Changelog • Documentation

This release brings the "custom metadata fields" feature. It allows you to define custom fields and associate values to your items. Additionally there are some ui and other fixes and improvements.

ℹ️ Existing users must enable "Custom Fields" in their users settings in order to use this feature.

• Feature: Custom Fields – define custom metadata fields an set values for them on your items. For example, this can be used to track invoice numbers, pagination stamps etc. Fields can be defined per collective and carry a format (or type). (#41)
• Feature: The language has been added to the metadata of an upload request and therefore overrides the collective's default language. This means you can now set the document language with each document. (#350)
• Feature: Show the currently logged in user and the collective in the web app. (#329)
• Feature: Tag categories are presented as a dropdown, where you can choose an existing one or type a new one. (#331)
• Feature: The dropdown fields for a person have been changed in that the options are now restricted to the corresponding scope: the correspondent person only shows persons not marked as concerning and vice-versa. (#332)
• Feature: Add CC and BCC recipients to item mail (#481)
• The consumedir.sh scripts was improved:
  • log a warning for all subfolders that currently wouldn't work due to configuration problems (#386)
  • ignore hidden files on linux (starting with a dot .)
  • include the parameter skipDuplicates into the upload request when the -m option is present
• Fixes a bug that prevented detecting dates in january (#480, thanks @vanto!)
• Fixes updating search view after changes like deleting item in multi-edit mode or updating tags via drag-and-drop.

The list of issues is here.

There were changes to the http endpoints; please checkout the Changelog for details.

Changelog • Documentation

This release adds the "preview feature", where a thumbnail of the first page of each attachment can be shown in the overview. Additional there are some bugfixes and UI improvements.

ℹ️ Docspell will start creating preview images for all your files when starting. If you don't see all of them immediatly, wait some time until they have been all created. It starts with newest files, so you should see some images shortly.

ℹ️ Existing users must (if wanted) enable the preview feature in the user ui settings.

• Feature: create a thumbnail of every file. The scale factor can be configured in the config file. The user can choose in the ui, what size to display and whether to display it or not. (#327)
• Feature: Display the number of pages in the overview (#325)
• Feature: Extend the upload request to allow to specify a file filter and a list of tags (#346, #385):
  • file filter: this is a glob that selects files when an archive file (a zip or e-mail file) is uploaded
  • tag list: allows to specify a list of tags that are automatically set on the item that is being created
  • these two settings are added to the scan-mailbox form and the source-form
• Feature: Filter mails to import by subject (#384)
• Feature: Add quick links to item-detail and overview to jump to a specific search view; e.g. click on correspondent takes you to the overview with this correspondent selected in the search menu (#355)
• Improved css assets (#349)
  • Moved from Semantic-UI to Fomantic-UI, which is an actively maintained fork of the former
  • Removed the request to obtain a google font. Now you can use docspell without hassle in environments without internet connection
  • jquery could be dropped as a js dependency
• Improved ui for multi select mode when selecting items
• Fix a bug when creating new scan-mailbox settings (#382)
• Fix a build issue that resulted in missing scripts in the tools zip file.
• Fix a bug that added the .pdf extension twice. The filename can now be defined in the config file. (#397)

There were changes to the http endpoints and additional config values; please checkout the Changelog for details.

Changelog • Documentation

This release contains many bug fixes, thank you all so much for helping out! There is also a new feature that allows you to edit multiple items at once; and more scripts in tools section:

• Edit/delete multiple items at once (#253, #412)
• Show/hide side menus via ui settings (#351)
• Adds two more scripts to the tools/ section (thanks to @totti4ever):
  • one script to import data from paperless (#358, #359), and
  • a script to check clean a directory from files that are already in docspell (#403)

It is also worth noting, that Steve Streza created a shortcut for iOS that allows to conveniently upload documents from your iPhone/iPad!

Please check out the Changelog or the milestone for a list of bugs fixed.

Changelog • Documentation

This release contains bug fixes and brings an improved docker setup. Thanks to all for providing great feedback and code!

• Improvements to the docker setup: application can be build from any version and is automated by dockerhub. Thanks to @totti4ever!
  • This change required breaking changes in the docker-compose.yml file. Please update your docker-compose.yml to the new version.
  • The image tags changed:
    • the -latest is now upper case, -LATEST
    • tagged releases include the version prefixed with a v, like in -v0.13.0
    • there are new -SNAPSHOT images that are build from the current master branch. Please note that snapshot versions may not be compatible with each other!
• The date extraction tried to create invalid dates (#298)
• Fix content column for MariaDB (#297)
• Fix retrying processing of files: attached files were not correctly found and the duplicate check must not run
• Reset upload page on init (#294)
• Fixes regarding base-url setting and auth cookie (#308)
• Fixes in openapi spec (#338, #343)
• Fixed error messages for modal dialogs (tag/organization/person) (#341)

Changelog • Documentation

This release focuses again on the web ui. Thank you all so much who contributed great feedback!

• Improve startup experience wrt the base-url setting. Docspell now uses request headers to determine absolute urls if this setting is unchanged. (#250)
• Fixes scrolling issues when going from detail to list view and vice versa. (#254)
• Fix quick search component to keep search term between changes. (#261)
• Docspell now provides a valid manifest to be an installable "pwa".
• Fixes consumedir.sh script to work with whitespace in filenames. (#269)
• Fix hidden fields feature. Hidden fields are now resetted when clicking Confirm. (#256)
• Improve Save Name in item detail: The save button is removed and changes are saved while typing. (#255)
• Add keyboard shortcuts in item detail for navigating and confirm/unconfirm. (#225)
• Detect duplicate files server side. The processing task is now able to detect duplicate files to skip them if desired. The ui has a new checkbox when uploading. (#252)
• Improve view of many attachments to an item: The tab menu is replaced by a dropdown menu that allows to change attachments also on small screens and/or when there are many attachments.

This is a bugfix release. For existing installations that use the full-text search index, a required schema upgrade was forgotten. It is included in this release.

New installations are not affected and both versions, 0.11.0 and 0.11.1 are otherwise identical.

Changelog • Documentation

This release contains improvements for document processing.

• Improves recognizing correspondents/concerned entities for new documents. It makes now better use of the complete data (address book) when analyzing documents.
• Adds text classification feature based on the stanford-nlp classifier: Docspell can now learn from your existing tagged documents using machine learning algorithms (by looking at the extracted text). For new documents, a tag of a given category is suggested. The learning task can be scheduled periodically. You need to enable it in the Collective Settings page.
• Adds support for French in text extraction and text analysis. The joex artifact size increases by about 100M, because of the included model files.

See the Changelog for more information.

Changelog • Documentation

This is now the 10th release of docspell! This time the focus is in the web ui, especially the card- or list view. Some "extra" for the 10th release is the new Android App that makes it convenient to upload files from your android devices. Together with a scanner app, like for example the free OpenNoteScanner you can quickly scan and upload receipts and other documents.

Other notable changes include:

• Rework of the search menu. Tags and folders are now presented as list. Drag & drop can be used in the list view to toggle tags and move items in folders (drag an item onto a tag or folder).
• Search by tag categories (#203)
• The sources page shows buttons to easier copy the urls as well as a QR code for using it with the android app.
• The search results contain part of the item notes which can be displayed in the card and configured in ui settings (#186, #201)
• Item notes in the detail view are reworked and always shown (#186, #192)
• Allows to edit metadata in the detail few (in contrast to just adding new ones), the modal now only covers the menu and not the document view (#205)
• Allow to hide some fields in the search and edit menus (#195)
• Reprocess files. Items can be submitted to be reprocessed. Also a task is provided to convert all non-converted pdfs using ocrmypdf introduced in the last release (#206). This can be used to convert all items created before 0.9.0 into pdfs with text layer.

Have fun!

Changelog • Documentation

This release adds the "folder feature" and contains some bug fixes.

• New feature: folders. Users can create folders and put items into them. Folders can have members (users of the collective) and search results are restricted to items that are in no folder or in a folder where current user is a member. (see #21)
• Folders can be given to the upload request.
• Add ocrmypdf utility to convert pdf->pdf with ocr-ed text layer.
• Extract PDF metadata and use the keywords to search for tags and apply them to the item during processing. (See #175)
• Several bug fixes

Some more can be found in the Changelog.

Changelog • Documentation

This release is all about full-text search.

• This release contains the full-text search feature (see #69). Full-text search is optional and is backed by Apache SOLR, it requires a running solr instance. It is disabled by default.

  The docker setup provides this component out of the box and therefore has full-text search enabled in the config file. Other setups need to take care manually (but it's not hard, just install solr and create a core).

• The consumedir.sh script was extended to use the integration endpoint for uploading. This makes it more convenient to use and allows to upload for all collectives.

Some more can be found in the Changelog. There were additions to the config files and the rest api.

Changelog • Documentation

This release focuses on the WebUI when working with items and documents. Notable changes:

• Enable paging in the item list view. When scrolling down, a button at the end can load more items.
• The item list contains all tags of an item.
• Tag colors can be customized.
• Item metadata can be added directly in the item detail view. (See #138)
• Allow for multiple notify-due-items tasks.
• Adds a "basic search" feature that looks in item notes, name and all names of correspondents/concerned enities.
• Allow to change names of attachments.

There is more, please see the Changelog for details. There were additions to the config files and the rest api.

Changelog • Documentation

This release includes a few features, some improvements and bug fixes. Notable changes:

• Import e-mails from your mailbox via periodic polling. You can now setup one or more tasks to run periodically that fetch mails from your mailbox and imports them into docspell.
• Add files to existing items.
• Change order of attachments
• Frontpage has been overhauled; items are presented as “cards” and the search menu can be moved away.
• Fix webapp for mobile devices.

There is more, please see the Changelog for details. There were additions to the config files and the rest api.

Changelog • Documentation

This release fixes various issues related to processing e-mail files. A new feature has been added allowing to be notified about due items.

Notable Changes:

• Notify about due items: A periodic task can be setup to check for upcoming due items and be notified via e-mail.
• Allow to delete single attachments from an item
• Many fixes when processing e-mail files

The configuration for the joex component has been changed. Please see the Changelog for more info.

Changelog • Documentation

This release adds support for zip and e-mail files and introduces a framework for running periodic tasks. While periodic tasks are currently only used internally, it prepares for user defined tasks coming in later releases.

Notable Changes:

• Support for archives files. Archives (currently zip and eml files) are extracted automatically and its content are added to an item. The major change is support for e-mail eml files. It allows to submit e-mails directly from your e-mail client into docspell.
• Improvements in text analysis.
• Fixes for handling text files.
• The pdf viewer can now be switched back to use the browsers builtin viewer. It seems that the viewerjs based pdf renderer cannot display all pdfs as good as the builtin viewer. Now you can choose.

The configuration for the joex component has been changed. Please see the Changelog for more info.

Changelog • Documentation

The third release focuses at opening Docspell to more file types and browsers.

• Support for many more document types has been added (including images and office documents). All input files are converted into PDF files and the original file is preserved.
• PDF Text extraction improved by omitting OCR if text can be stripped.
• There is a new PDF viewer (utilizing viewerjs) that also works in mobile browsers.
• Improve editing notes: Since notes may evolve, there is now a larger edit form and a markdown preview.
• Show the extracted information (text, labels, proposals) of an attachment in the Webui.
• The name search now also searches in item notes.
• Bug fixed where it was possible to create invalid input when creating new sources.
• Bug fixed where the item menu was not properly initialized for equipments.
• The ds.sh script has now an option to check a file for existence in docspell.

This introduces two more external tools to the joex component that do the heavy lifting: unoconv and wkhtmltopdf. The configuration for the joex component has been changed. Please see the Changelog for more info.

The second release of Docspell addresses some annoying issues in the UI and adds a "send by email" feature.

Notable Changes:

• Send an item and its attachments via E-Mail (requires to setup SMTP settings per user)
• Add a search field for meta data
• The item detail view is now a perma-link
• New endpoints to check whether a file is in Docspell by using their SHA-256 checksum (see the api doc here and here), the scripts in tools/ now use this endpoint to skip existing files
• Better support multiple attachments with long names in the UI
• Fixes textarea updating issues

Checkout the updated documentation for more details.

The initial release of Docspell containing the basic features with a Web UI:

• Create items by uploading PDF files
• Analyze the PDF files and propose meta data
• Manage meta data and items
• View processing queue

See the microsite for documentation.