org-roam-id-get-file no longer falls back onto the current buffer: It queries the org-roam database, and optionally org-id-locations, and declares that it does not exist if it isn't in either.
Instead of maintaining a file queue to process for updating the Org-roam
database, we instead simply call `org-roam-db-build-cache` to rebuild
the db. `org-roam-db-build-cache` feels Fast Enough(TM), and basically
runs instantly if no files are modified. This greatly simplifies the
code, basically allowing to maintaining a single code path for db
operations.
This PR should also address the slowness wrt to ID links. Org-roam now
builds the headline information first, so there is no need to read the
file to check if the headline is there. The downside to this is that
this only works for IDs in Org-roam files.
* (fix): use org-roam-link-title-format in link replacement
Makes `org-roam-format-link` respect `org-roam-link-title-format`.
Also removes lowercasing via prefix argument in `org-roam-insert`, not
sure how many people use it.
* update changelog
* (fix): fall back to org-id search ID face computation
Fixes the scenario where the face is reported as invalid although it is
part of Org's ID files (e.g. in the agenda).
Fixes#1191.
This can however slow face computation a lot (when there are many org
id/agenda files). Here, we choose to focus on correctness.
* Document faces, default to only applying to Org-roam notes
Otherwise it breaks `org-roam-db-build-cache` as it relies on temporary buffer.
Without keeping the buffer, `org-roam--extract-links` kills it and any further
functionality of file processing breaks.
Prevent error when trying to move file to same name
Prevent string matching in org from messing with the current re-search-forward
position which has caused infinite broken search
Additionally cache IDs at outline-level 0, now that property drawers are supported in Org 9.4.
Update org-roam--format-link to prefer ID links wherever possible. That is, when a file has an ID, use an id link instead of file link.
Adds `org-roam-link-file-path-type`, used for link path computation
wherever sensible. This includes in the org-roam backlinks buffer, and
in link replacement.
Also moves link expansion/fixing from cache build to Org-roam buffer
render time. This reduces cache build time, but makes buffer rendering
slightly slower.
Previously we had overwritten the Org's store link function for files,
to create IDs if the point was under the headline. This has several
issues:
1. It becomes impossible to store a link to the file using `org-store-link`
2. This override is global across Org, although we had a guard in
`org-roam-store-link`
3. IDs are created during an org-capture, because org-capture calls `org-store-link`
This is unnecessary. Org provides `org-id-store-link` instead for this
purpose. Instead, we advice `org-id-new`, to queue the file for a DB
update if an ID was created.
`find-file-noselect` is affected by the user's init.el, and there is
little control over what is being activated. For database updates, read
access is all we need, so we use `with-temp-file`.
This changes ref extraction to parse the ref using the org plain-links
syntax. This works for arbitrary `cite` links, and web links.
Link storage and ref lookup is now based on the path (e.g.
`//google.com`) or `author_year`.
file-truename calls alone accounts for over 20% of CPU samples in org-roam-db-build-cache. It's expensive and unnecessary. To get an absolute path, expand-file-name alone is enough.
Co-authored-by: Jethro Kuan <jethrokuan95@gmail.com>
PR #1076 prevented the title change hook from causing errors for new files,
but completely disables the behaviour. To cause the title hooks to take
effect, we set the local variable `org-roam-current-title` upon save,
regardless of whether the hook ran.
This should have the effect of enabling the title change behaviour upon
first save, assuming the save succeeds.
Addresses https://org-roam.discourse.group/t/org-roam-update-file-name-on-title-change-works-discontinuously/767
Attempting to perform link expansion on URL links caused extreme slowdowns in Windows environment. Ref #1038#1067.
Co-authored-by: Noboru Ota <me@nobiot.com>
Org expects face functions to save-match-data, or it would lead to
infinite recursions with use of some Org functionality such as
org-agenda (h.t. @myshevchuk for figuring this out)
Fixes#1096 and fixes#1045.
`org-roam-enable-fuzzy-links` controls whether to enable Org-roam's
fuzzy link behaviour. This can be undesirable to some users, who would
like to use [[foo]] links to reference named blocks
Fixes#1082, #1077, #1074
When reverting link extraction to using the `org-element` api, the
preview content (org-element-at-point) was incorrectly extracted,
causing incorrect preview content and cryptic messages.
org-element-at-point doesn't just determine the closest element -- if
it's "within blank lines at the beginning of buffer", it returns nil.
;; Within blank lines at the beginning of buffer, return nil.
((bobp) nil)
Skip whitespace at beginning of buffer to avoid this.
Adds `org-roam-title-change-hook`. This now contains two functions:
1. `org-roam--update-links-on-title-change`: updates the link
description of all files that link to the current file.
2. `org-roam--update-file-name-on-title-change`: updates the current
buffer's filename to reflect the updated title.
Previously the rename file advice was responsible for trying to update
the link descriptions as well. This was brittle and didn't work in some
cases.
The rename-file-advice has now been altered to solely deal with breaking
links. We use the before and after-save-hooks to determine whether the
title has changed, and update the link descriptions accordingly.
Fuzzy links were initially detected as anything within double brackets.
This could include code in source blocks.
This PR introduces `(org-roam--fuzzy-link-p)`. This uses an additional
check using the `org-element` API, and ensures that the link type is
fuzzy (not a file: or https: link, for example).
Fixes#1069, and an array of unreported bugs:
1. Link extraction into the database should no longer pick up false
links (in code blocks, for example).
2. Link completion will only truly work within fuzzy links
