Commit graph

1401 commits

Author SHA1 Message Date
Debanjum Singh Solanky
013c7c10a4 Bump khoj pre-release version 2023-01-22 18:45:56 -03:00
Debanjum Singh Solanky
ad3c9b5f44 Bump khoj version to 0.2.5 in preparation for release 2023-01-22 18:18:21 -03:00
Debanjum Singh Solanky
9ed056c7e7 Use consistent indentation in Khoj Emacs Readme 2023-01-22 18:04:12 -03:00
Debanjum Singh Solanky
0980c6e87f Update Emacs Usage section in Readme. Add find-similar, menu usage 2023-01-22 18:04:12 -03:00
Debanjum Singh Solanky
6908b6eed3 Truncate image queries below max tokens length supported by ML model
This would previously return the infamous tensor size mismatch error
Verify this error is not raised since adding the query truncation logic
2023-01-21 14:11:00 -03:00
Debanjum Singh Solanky
3d9ed91e42 Search by image at path only if query of form "file:/path/to/image"
Previously no query syntax helpers, like the "file:" prefix, were used
before checking if query contains file path.

This made query to image search brittle to misinterpretation and
pointless checking

Add test to verify search by image at file works as expected
2023-01-21 14:06:56 -03:00
Debanjum
655ef11653
Find Similar Notes, Transactions, Images from Khoj in Emacs
### Overview
Find items of specified type similar to current text item at point

### Capabilities
- Support querying with text surrounding point in any text buffer
- Find similar items of specified content type indexed on Khoj

### Details
- Query using text in current section if in a `outline-mode` buffer (i.e markdown heading, org-mode entry text)
- Query using text in current paragraph if in non `outline-mode` buffer
- Search for items of `content-type` set in khoj transient menu
- Update last used khoj content-type and results from the
  *find-similar* and *update* functions for later reuse

### Related
- Recently added [Find Similar Notes in Khoj Obsidian](https://github.com/debanjum/khoj/pull/122) as well
2023-01-20 22:44:28 -03:00
Debanjum Singh Solanky
b7aa22a059 Change order of arg passed to query-api-and-render-results by importance 2023-01-20 22:13:24 -03:00
Debanjum Singh Solanky
936a88fa7e Find items of specified type similar to current text item at point
- Support querying with text surrounding point in any text buffer
  Previously could only find items similar to org entry at point

- Find similar items of specified content type indexed on khoj
  Previously only looked for similar org entries indexed on khoj

  Now uses the content-type configured in khoj transient menu to find
  items of the specified content type

- Details
  - Generalize the get-current-org-entry-text func to get text for any
    outline section
  - Replace leading whitespaces from query text as well
  - Create method to get current paragraph text from non-outline mode
    buffers
  - Update transient, find-similar funcs to pass, use content-type
    configured in khoj transient menu
  - Generalize query title creation logic to remove markdown headings
    prefix (#) apart from org heading prefix (*) as well
  - Update last used khoj content-type and results from the
    find-similar and update funcs for later reuse
  - Jump to top of results buffer after results rendered
2023-01-20 22:12:54 -03:00
Debanjum Singh Solanky
17aaadea1f Find notes similar to current org entry at point 2023-01-20 05:14:54 -03:00
Debanjum Singh Solanky
44bbc0a417 Add section separators to khoj.el for easier code traversal 2023-01-19 23:36:54 -03:00
Debanjum
7516435a0b
Automate khoj.el build and quality checks
- 9f0bd0a Build `khoj.el' and Run `package-lint', `checkdoc' and other melpa package quality checks
- 48ad3c5 Use default content types if fail to call backend on `khoj.el` load
2023-01-19 20:21:55 -03:00
Debanjum Singh Solanky
48ad3c535e Use default content types if fail to call backend on khoj.el load
Do not want khoj.el to fail on init/load if khoj backend not running
2023-01-19 20:13:49 -03:00
Debanjum Singh Solanky
9f0bd0a361 Add Github workflow for khoj.el build and quality checks
Add khoj.el build badge to khoj.el Readme
2023-01-19 20:13:19 -03:00
Debanjum
b58dd82141
Use Transient Menu to Improve Khoj.el Interface
- 5f446b1 Convert `khoj' entry point method to transient.el menu for richer configuration
- 9d64a00 Allow updating khoj content index from within `khoj.el'
2023-01-19 03:11:23 -03:00
Debanjum Singh Solanky
0dd1cba272 Rename configuration sections in khoj.el transient menu 2023-01-19 03:03:08 -03:00
Debanjum Singh Solanky
5d0f369186 Add ability to quit khoj transient with standard q keybinding 2023-01-19 02:47:07 -03:00
Debanjum Singh Solanky
87c7cf4272 Use single khoj func as entrypoint. Group khoj.el code into sections
- Give more relevant, specific name to khoj suffix commands
- Remove `khoj-simple'. Have single `khoj' function for entrypoint
2023-01-19 02:38:19 -03:00
Debanjum Singh Solanky
9d64a009fd Allow updating khoj content index from within khoj.el
- Split transient config menu by type
2023-01-18 23:07:59 -03:00
Debanjum Singh Solanky
a8d0c7d905 Rename search type to more apt content type in khoj.el 2023-01-18 22:13:49 -03:00
Debanjum Singh Solanky
00daea16df Allow setting default-search-type to image. Make docstrings compact 2023-01-18 22:01:17 -03:00
Debanjum Singh Solanky
216b17cfd0 Dynamically populate content type choices when khoj transient invoked 2023-01-18 22:00:56 -03:00
Debanjum Singh Solanky
5f446b1440 Convert main khoj.el entrypoint into transient menu for richer configuration 2023-01-18 21:50:07 -03:00
Debanjum Singh Solanky
5c07dcd219 Fix, update Obsidian Readme. Add Find Similar Notes to Implementation section 2023-01-18 00:22:26 -03:00
Debanjum
b7fc344be1
Search for Similar Notes from Obsidian Plugin
Enable searching for notes similar to the current note being viewed

## Main Changes
- 39a18e2 Extend search modal to search for similar notes
  - Hide input field on init, Trigger search on opening modal when in similar notes mode
  - Set input to contents of current markdown file and get notes similar to it
  - Re-rank, by default, when searching for similar notes
  - Filter out current note from similar note search results
- 0bed410 Only show `Find Similar Note' command in Editor
2023-01-18 00:10:10 -03:00
Debanjum Singh Solanky
6119d0a69e Add usage of "Find Similar Notes" command to the Khoj Obsidian Readme 2023-01-18 00:03:13 -03:00
Debanjum Singh Solanky
657e455785 Remove unused `onunload' method in main.ts of khoj obsidian plugin 2023-01-17 23:46:38 -03:00
Debanjum Singh Solanky
0bed410712 Limit Find Similar Note command to be triggered from Editor
Fixup indentation and comments
2023-01-17 19:34:48 -03:00
Debanjum Singh Solanky
39a18e2080 Add ability to search for similar notes in Khoj Obsidian
- Hide input field on init, Trigger search on opening modal in similar notes mode
- Set input to current markdown file and get similar notes to it
- Enable rerank when searching for similar notes
- Filter out current note from similar note search results
2023-01-17 19:07:18 -03:00
Debanjum Singh Solanky
ffaef92476 Encode query string before passing as query param to search API 2023-01-17 18:04:11 -03:00
Debanjum Singh Solanky
d5a7cc5b0f Compact code to map results from search API into SearchResult objects
Make code compact for readability
Remove unneeded temporary variables and return statements
2023-01-17 18:04:11 -03:00
Debanjum Singh Solanky
8ab7a26bde Update Khoj on Obsidian screenshots in Main and Plugin Readme
- Screenshot querying "Setup Editor" on test vault with Khoj Readmes
- New features showcase:
  - information keybindings, rerank keybinding at bottom of modal
  - fixed top level headings in search results
  - search results snipped if greater than N words
2023-01-17 13:58:50 -03:00
Debanjum Singh Solanky
7b4f78776c Fix extracting Markdown Entries with Top Level Headings
- Previously top level headings would have get stripped of the
  space between heading text and the prefix # symbols. That is,
  `# Top Level Heading' would get converted to `#Top Level Heading'
- This would mess up their rendering as a heading in search results

- Add unit tests to text_to_jsonl processors to prevent regression
2023-01-17 13:06:28 -03:00
Debanjum Singh Solanky
1a296518c5 Limit total words for each Search Result rendered in search modal
Provides a more consistent rendering of results in modal.
Makes it easier to see more results in modal.
To see complete entry, user can always just jump to entry from modal
2023-01-17 13:06:14 -03:00
Debanjum Singh Solanky
e7b89f7fd0 Return compiled entry in additional details of /api/search response
This can be used to highlight portion of raw entry to highlight and
for passing to summarizer to stay with max_tokens limit supported by
GPT models
2023-01-16 22:56:06 -03:00
Debanjum Singh Solanky
7071d081e9 Increase max_tokens returned by GPT summarizer. Remove default params 2023-01-16 22:55:36 -03:00
Debanjum Singh Solanky
3d9cdadbbb Add codebase visualization of Khoj Obsidian to Khoj Obsidian Readme 2023-01-15 14:09:21 -03:00
Debanjum Singh Solanky
d02ba325aa Handle empty chat history returned by API to chat.html on web interface 2023-01-15 13:51:16 -03:00
Debanjum Singh Solanky
721bbbe15c Update Readme. Add Chat with Notes Section to Advanced Usage
- Add Setup OpenAI API key in Khoj Section to Miscellaneous
  Refer all mentions of setting up your OpenAI API key to that section
- Add Demo Screenshot of Chat with Notes
- Put existing Miscellaneous Section under Beta API sub heading
- Fix to make Access Khoj on Mobile a Subsection of Advanceed Usage

- Trigger refresh of github image cache by adding ? at end of image paths
2023-01-14 00:39:15 -03:00
Debanjum Singh Solanky
42f8230b37 Update Troubleshooting Section in Main Readme
- Convert Troubleshooting Issues into Headings instead of Bullets
  Allows them to be linked to more easily. E.g when pointing folks to
  it in github issues etc

- Add index corruption issue and fix to the Troubleshooting section
2023-01-13 23:03:15 -03:00
Debanjum
3f2ea039a7
Add Chat page to the Khoj Web Interface
### Overview
- Provide a chat interface to engage with and inquire your notes
- Simplify interacting with the beta `chat` and `summarize` APIs

### Use
- Open `<khoj-url>/chat`, by default at http://localhost:8000/chat?type=summarize
- Type your queries, see summarized response by Khoj from your notes

**Note**:
 - **You will need to add an API key from OpenAI to your khoj.yml**
 - **Your query and top note from search result will be sent to OpenAI for processing**

## Details
- 177756b Show chat history on loading chat page on web interface
- d8ee0f0 Save chat history to disk for persistence, seeing chat logs
- 5294693 Style chat messages as speech bubbles
- d170747 Add khoj web interface and chat styling to new chat page on khoj web
- de6c146 Implement functional, unstyled chat page for khoj web interface
2023-01-13 23:02:19 -03:00
Debanjum Singh Solanky
16d4560ff8 Comment css styling of chat page for later reference 2023-01-13 22:40:01 -03:00
Debanjum Singh Solanky
cfef346d03 Do not update query field to ever chat message
It doesn't work as well with chat, unlike for search page
Use more appropriate thinking face emoji for you instead of surprise face
2023-01-13 22:24:26 -03:00
Debanjum Singh Solanky
177756be7e Fetch chat history from backend and render it on chat page load 2023-01-13 22:01:57 -03:00
Debanjum Singh Solanky
330febaa1a Update conversation logs from /beta/summary API endpoint too 2023-01-13 22:01:57 -03:00
Debanjum Singh Solanky
cb6f0b53c9 Make user_message_metadata arg to message_to_log in gpt.py optional
- Use a default user_message_metadata if arg not set
- Update conversation to use `by' as `you' and `khoj'
2023-01-13 22:01:57 -03:00
Debanjum Singh Solanky
cc2456e411 Update /beta/chat API to return chat history if no query param passed 2023-01-13 22:01:57 -03:00
Debanjum Singh Solanky
d8ee0f0e9a Use scheduler to save chat history to disk every 5 minutes
- The previous mechanism to trigger saving on shutdown event did not work
- Use scheduler to persist chat sessions to disk at a 5 minute interval
  - This improve time granularity, fixed interval of saving chat logs
  - It may lose ~5 minutes of chat history until mechanism to also
    write on shutdown found/resolved
- Create conversation directory if it doesn't exist before attempting write
- Reset chat_session after writing it to disk
2023-01-13 22:01:57 -03:00
Debanjum Singh Solanky
5294693e97 Style message as speech bubbles on chat page of web interface
- Wrap messages into speech bubbles
  - Color messages by khoj blue, sender grey
  - Add those standard protrusions to the speech bubbles for fun

- Align bubbles left or right based on sender
  - messages by khoj are left aligned, message by self are right aligned

- Put message metadata like sender and time under speech bubble
  - use data-* attribute and ::after css pseudo-selector for this

- Update renderMessage func to accept time param, remove unused type_ param
2023-01-13 22:01:57 -03:00
Debanjum Singh Solanky
7723d656dc Do not force GPT to summarize note using past tense
Not all notes are in the past. Notes can be about stuff in the future.
Casting them to past tense gives the impression that they've already
happened / been done.
2023-01-13 13:10:35 -03:00