aboutsummaryrefslogtreecommitdiffhomepage
path: root/ide/sentence.ml
Commit message (Collapse)AuthorAge
* Fixing ml-doc.Gravatar Pierre-Marie Pédrot2014-05-01
|
* CoqIDE: error reporting fixedGravatar gareuselesinge2013-10-10
| | | | | | | | | Many things were wrong. Error tags were deleted by mistake, the screen was recentered on `INSERT using the wrong function (that cause some horizontal scrolling even if it was not needed), the cursor not advanced to the end of the wrong sentence. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@16874 85f007b7-540e-0410-9357-904b9bb8a0f7
* CoqIDE: a comment is not a sentenceGravatar gareuselesinge2013-10-10
| | | | | | | | | | | | | | | | | | | | | This simplifies the whole document business: the document on the Coq side has the very same nodes as the CoqIDE document, there are no "fake" nodes in the CoqIDE document to be skipped over. We keep the comment tag stamped by the coq_lexer module, since we may want to allow edits in there without telling Coq (as proof general does). Not implemented yet, but doable thanks to the comment tag. Pierre Boutillier suggested that this makes back-1-sentence ugly, since it moves the cursor far away if the sentence begins with a comment. While this is true, *today*, there is no need to undo the last sentence with the button to edit the text. One can just move the cursor where he likes and edit. In this case the sentence is backtracked automatically and the cursor is left where it is. Hence considering initial comments as part of the following sentence should not be an usability issue anymore. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@16866 85f007b7-540e-0410-9357-904b9bb8a0f7
* CoqIDE ported to the revides protocolGravatar gareuselesinge2013-09-30
| | | | | | | | | | | - the zone to be edited is now between the marks start_of_input and stop_of_input - when -debug is given, the edit zone is underlined - the cmd_stack is focused/unfocused according to the new protocol - read only tag resurrected and used to block the "Qed" ending a focused zone git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@16814 85f007b7-540e-0410-9357-904b9bb8a0f7
* Coq_lex: direct accounting of utf8 extra bytes in offsetsGravatar letouzey2012-12-11
| | | | | | | | | We directly produce in Coq_lex a utf8 char offset instead of a byte offset, by counting the utf8 extra byte during the lexing. This way, no need anymore for converting later with complex byte_offset_to_char_offset. git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@16059 85f007b7-540e-0410-9357-904b9bb8a0f7
* Coqide: restore the tag removal of copy-pasted zonesGravatar letouzey2012-12-10
| | | | | | | | The handler for apply_tag removed in commit 16044 was probaly meant for that. We now proceed in a more simple way, in Sentence.split_slice_lax, instead of doing a remove_tag in a apply_tag handler (!). git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@16058 85f007b7-540e-0410-9357-904b9bb8a0f7
* Coqide: some more refactoring to lighten coqide.mlGravatar letouzey2012-12-10
Main victim is analyzed_view : - some unnecessary methods have been killed (hep_for_keyword for instance) - some other migrated elsewhere (recenter_input, find_next_occurrence, ...) - analyzed_view is now split in two : fileops (filename, save, revert, ...) and coqops (process_next_phrase, ...) Four new files created: - Sentence (for tag_on_insert and alii) - FileOps (ex-first-half of analyzed_view) - CoqOps (ex-second-half of analyzed_view) - Session (ex-record viewable_script and functions about it) Also lots of renaming, trying to be shorter (but still meaningful) and more uniform git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@16057 85f007b7-540e-0410-9357-904b9bb8a0f7