TheAnachronism/docspell
1
0
Fork 0
mirror of https://github.com/TheAnachronism/docspell.git synced 2024-11-13 02:31:10 +00:00
Code Issues Packages Projects Releases Wiki Activity
NaN Commits 2 Branches 1 Tag 89 MiB
5f32ca4cdc
Commit Graph

124 Commits

Author SHA1 Message Date
Eike Kettner
ab1139523a Let the convert-all task retry when pdf conversion fails 2020-10-26 23:39:26 +01:00
Eike Kettner
b59696a9d3 Make sure to only remove/retry items in premature states 2020-10-26 23:39:26 +01:00
Eike Kettner
26e89bf84e Edit org/person/equipment of multiple items 2020-10-26 13:35:47 +01:00
Eike Kettner
2e6026b817 Edit dates of multiple items 2020-10-26 13:16:03 +01:00
Eike Kettner
3c0b86cb19 Fix regex patterns used for NER 
Patterns are split on whitespace by the nlp library and then compiled,
so each "word" must be a valid regex.

Fixes: #356
 2020-10-21 00:55:14 +02:00
Eike Kettner
3f697f51aa Autoformat 2020-10-06 23:31:09 +02:00
Eike Kettner
d4354b8b49 Skip pdf conversion if a converted file exists 
For images the conversion also returns the extracted text. If this
would have failed to be saved, it is extracted in the following
text-extraction step.
 2020-10-02 17:39:39 +02:00
Eike Kettner
b6f23b038a Fix finding attachments for retries 
The attachments to process again must be searched in sources and
archives, too.
 2020-10-02 17:39:34 +02:00
Eike Kettner
5e21552358 Don't do duplicate check on retries 2020-10-02 16:50:52 +02:00
Eike Kettner
f6f63000be Prepend a duplicate check when uploading files 2020-09-23 23:37:00 +02:00
Eike Kettner
c658677032 Autoformat 2020-09-09 00:29:32 +02:00
Eike Kettner
76ccfb8a81 Only learn from confirmed items 
Text classification should only learn from confirmed items. Log if
classification is disabled when processing an item.
 2020-09-07 13:04:40 +02:00
Eike Kettner
4309bd8dfd Some cleanup 2020-09-02 21:22:30 +02:00
Eike Kettner
237b960625 Guess a tag on item processing using a trained model if available 2020-09-02 18:28:14 +02:00
Eike Kettner
316b490008 Implement learning a text classifier from collective data 2020-09-02 18:28:14 +02:00
Eike Kettner
68bb65572b Integrate learn-classifier task into the app 2020-09-02 18:28:14 +02:00
Eike Kettner
0c97b4ef76 Initial impl of a text classifier based on stanford-nlp 2020-09-02 18:28:14 +02:00
Eike Kettner
8c4f2e702b Add classifier settings 2020-09-02 18:28:14 +02:00
Eike Kettner
3473cbb773 Use collective data with NER annotation 2020-08-25 20:40:44 +02:00
Eike Kettner
96d2f948f2 Use collective's addressbook to configure regexner 2020-08-24 14:40:52 +02:00
Eike Kettner
8628a0a8b3 Allow configuring stanford-ner and cache based on collective 2020-08-24 10:55:59 +02:00
Eike Kettner
3986487f11 Add api docs and cleanup 2020-08-13 21:22:54 +02:00
Eike Kettner
41ea071555 Add a task to convert all pdfs that have not been converted 2020-08-13 01:06:13 +02:00
Eike Kettner
07e9a9767e Add a task to re-process files of an item 2020-08-12 22:29:56 +02:00
Eike Kettner
09d74b7e80 Return item notes with search results 
In order to not make the response very large, a admin can define a
limit on how much to return.
 2020-08-05 00:09:37 +02:00
Eike Kettner
45b0deeced Print solr url on start 
This is useful info to see which url has been selected, same as db
connection.
 2020-08-01 15:59:14 +02:00
Eike Kettner
1fc57fc2b2 Set default value for min-text-len to 500 
This value is used to decide whether to try OCR or not. If text is
below this value, OCR is run and both results are compared. It was set
to 10, which is just one or two words. Since the context for docspell
are documents, this value is too low.
 2020-08-01 15:46:00 +02:00
Eike Kettner
cec4948710 Add pdf meta data to extracted text to add it to full-text index 2020-07-19 01:07:49 +02:00
Eike Kettner
209c068436 Use keywords in pdfs to search for existing tags 
During processing, keywords stored in PDF metadata are used to look
them up in the tag database and associate any existing tags to the
item.

See #175
 2020-07-19 00:28:04 +02:00
Eike Kettner
bd20165d1a Use given folder-id when adding initial fts docs 2020-07-18 23:04:01 +02:00
Eike Kettner
3d49ceaab5 Use ocrmypdf tool to create pdf/a during conversion 
- Use another external tool to convert pdf to pdf which also adds the
  extracted text as another layer into the pdf

- Although not used, the external conversion routine will now check
  for an existing text file that is named as the pdf file with extension
  `.txt`. If present it is included in the conversion result and will be
  used as the extracted text.

- text extraction for pdf files happens now on the converted file,
  because it may already contain the text from the conversion step and
  thus avoids running OCR twice.

- All errors during conversion are not fatal; processing continues
  without a converted file.
 2020-07-18 17:19:29 +02:00
Eike Kettner
5b01c93711 Add a folder-id to item processing 
This allows to define a folder when uploading files. All generated
items are associated to this folder on creation.
 2020-07-14 23:18:39 +02:00
Eike Kettner
259526a088 Organize imports 2020-07-12 13:51:52 +02:00
Eike Kettner
22fa1dba13 Apply folder restriction to fulltext only search 
And update index when folder changes.
 2020-07-12 13:50:45 +02:00
Eike Kettner
aeba4ba913 Refactor full-text migrations and add folder to solr schema 2020-07-12 13:50:14 +02:00
Eike Kettner
e387b5513f Remove items in non-member folders from sql search results 2020-07-11 22:25:56 +02:00
Eike Kettner
752a94a9e2 Implement space operations 2020-07-11 01:30:28 +02:00
Eike Kettner
347a029af8 Scalafix organize-imports 2020-06-28 21:20:47 +02:00
Eike Kettner
41c0f70d3b Fix cancelling jobs 
A request to cancel a job was not processed correctly. The cancelling
routine of a task must run, regardless of the (non-final) state. Now
it works like this: if a job is currently running, it is interrupted
and its cancel routine is invoked. It then enters "cancelled" state.
If it is stuck, it is loaded and only its cancel routine is run. If it
is in a final state or waiting, it is removed from the queue.
 2020-06-26 23:08:27 +02:00
Eike Kettner
d79ae6233a Restrict proposals for due date 
Avoid dates too far in the future.
 2020-06-26 16:58:17 +02:00
Eike Kettner
91da3b149e Reducing default retries to 2 
Many errors cannot be recovered from by retrying. There is currently
no way to distinguish these states so it is now set to a lower value
to have not long wait times until an item arrives.
 2020-06-25 23:57:01 +02:00
Eike Kettner
dc8f1a0387 Fix global re-index task to re-create the schema 
Otherwise new instances could not be re-indexed.
 2020-06-25 23:02:06 +02:00
Eike Kettner
14213c4c27 Allow some solr query options in the config file 2020-06-24 23:37:20 +02:00
Eike Kettner
532caed84c Consistent logging of request/responses to solr 
Using a middleware. Also add missing changesets for mariadb.
 2020-06-24 21:25:46 +02:00
Eike Kettner
47697a8056 Set some logs to trace 2020-06-24 01:16:13 +02:00
Eike Kettner
e06a3f8fdd ScalafmtAll 2020-06-23 00:18:59 +02:00
Eike Kettner
ffbb16db45 Transport highlighting information to the client 2020-06-23 00:17:29 +02:00
Eike Kettner
cfe5aa8894 Use no-op fts-client if disabled + push this flag to the webui 2020-06-21 21:06:08 +02:00
Eike Kettner
0d8b03fc61 Add backend operations for re-creating the full-text index 2020-06-21 15:46:51 +02:00
Eike Kettner
14ea4091c4 Renaming things 2020-06-21 13:15:02 +02:00