This was already done for the goto table, but by first looking up whether a label already exists in the table. Since the insert operation already reports entries with the same key in the tree, we should rather check that. This saves one goto table lookup per label definition, but on the other hand results in excess allocations in case of encountering a label twice. In teco_help_set(), we did not check for collisions at all. Even thouch they are very unlikely, we could encounter them if a topic is defined on exactly the same position as in .SCITECO_TOPIC foo .SCITECO_TOPIC foo In this case, we leaked memory.
* This mirrors how the `FT$` command toggles through several results. * Some non-unique .SCITECO_TOPICs have been added to sciteco(1) and sciteco(7). * This works by inserting non-unique keys into the help topic RB tree - or rather by turning filename:pos also into keys since rb3ptr does not allow non-unique keys. When searching for a topic, we sort lower than any other entry with the same name. This way we can easily iterate all entries with the same name (search topic). The alternative would have been to maintain a linked list in each and every teco_help_topic_t. * teco_rb3str_auto_complete() now filters out possible duplicate keys. * Updated test case: check for non-unique topic names.
fixup to 5b9ac0d43650bc7a1fe5d6e134e74bda1243fd99.
That variant of the command is explicitly used to launch the LSP. It makes no sense to return an error message in case of success.
This currently only works in insert commands after enabling the immediate editing modifier via CTRL+G and if the LSP was already launched (e.g. by an explicit `FT$`). File name completions are still performed with an activated ^G modifier only if the LSP is shut down. Just like for file names, the modifier is automatically cancelled after a successful completion - there is little sense in completing several symbols in a row.
This will be important once we're doing LSP auto-completions. Until a completion is complete, we do NOT want to disable the modifier however since otherwise it can be disabled after a partial completion if there happened to be a rubbed out command line. Discarding a rubbed out command line consequently no longer disables the modifier as well. Any change of the modifier is now logged with an Info-level message. Some rudimentary test case has been added for rub-in via ^G.
This is similar to `\` at the line end in many languages like C and allows you to split a very long string argument over several lines without adding the EOL characters themselves. The TECO equivalent to `\` would be ^Q, but it would be a semantic exception since ^Q (escaping) is supposed to _preserve_ the value of the following character. Therefore I decided to use ^E instead. This works with DOS and Mac line breaks as well.
Previously we were using both Scintilla undo actions and our own undo tokens and both had to be coordinated via `undo__teco_interface_ssm(SCI_UNDO, 0, 0)`. This was error prone since we have to predict when an undo action is actually generated. Furthermore, you had to update SCI_SETUNDOCOLLECTION on all buffers and the Q-Reg view whenever switching to and from interactive mode. Also, this probably wasted memory for the doubled undo bookkeeping. Now, we always set SCI_SETUNDOCOLLECTION(FALSE), even in interactive mode. teco_undo_view_insert() was introduced to undo text deletions. While this saves almost no lines of code, it's just a clearer and simpler architecture.
This can be read by clangd as an alternative to compile_commands.json (which would be trickier to generate with Autotools). However, since this is generated in the build directory, it may not always be picked up automatically -- a build/ directory should work, but otherwise you might have to pass --compile-commands-dir to clangd.
* These are defined by C23 and jemalloc actually provides them beginning with FreeBSD 16. Since glib is using free_sized() when available, builds were broken on FreeBSD 16. * Should be backported to the SciTECO 2.5.2 FreeBSD port to silence the fallout.