Try to correct these regressions. Note that the iterator-part may not be ideal; I understand that removing iterators from the start of simple name lookup was a deliberate decision and I don't want to revert all of that. Thus the new text is somewhat awkward. Closes #2175
Improve so that .13 is a floating point number, (#2491)
Closes #2122 * Improve so that .13 is a floating point number, and some minor corrections. * Introduce UNSIGNED-REAL and UNSIGNED-NUMBER to be able say that 13 is not a floating point, but can still be used everywhere where numbers are expected (13. is a floating point number). * Add more examples * Simplify descriptive text as proposed with good link to grammar
* Clarify needed defaults for old modifiers. Closes #2517 * Clarify new modifiers Closes #2519 * Clarify when to use convertElement and convertClass Closes #2520 * Clarify goal of conversion Closes #2514 * Add examples (including base-class variants)
This fixes most of #2364 (carefully avoiding wording that would trigger GitHub action).
Summary of what this PR aims to cover: - No upright quotes should appear outside of code snippets (`lstlisting` and `lstinline`). - No quotes around inline code. - Use `\lstinline` instead of quotes for inline code. - Use `\emph` when there is a need to remind that something has a special meaning in the context of the specification (when the thing was introduced, it should have been marked with `\firstuse`, see below). - Titles of publications, chapters, and sections in `\emph` instead of quotes. - Use matching pairs of double quotes (“conceptual”) when using a word or phrase that is intentionally vague, lacking proper definition, or otherwise calls for the reader's imagination. - Single quotes around single special characters where needed to get more clear separation from surrounding text. Example (somewhat damaged by GitHub's font): > …there is an operator ‘`0`’ defining the zero-value - Filenames in `\filename` (pretty big risk of not having found all occurrences). - Use of `\firstuse` instead of `\emph` when a term is introduced (pretty big risk of not having found all occurrences). - Use of the `example` and `nonnormative` environments whenever possible with at most minimal changes to content, in order to avoid misuse of interleaved fragments in `\emph`, making the source code maintainable, avoiding lots of tiny errors in the markup, and correcting many places where formatting was missing between the `\emph` fragments.
Parts of #2364 not covered by this PR: - Proper handling of references. - Cleanup inside non-normative text that couldn't be converted to use the `nonnormative` environment due to not being a stand alone sequence of paragraphs. This requires #2368.
Minor changes: * Clean up use of ``...'' * Clean up related to " being used outside of code snippets * Use \filename to markup some filenames * Get rid of some meaningless uses of \textbf * Improve markup in simple cases of non-normative text and examples * More cleanup, with special attention to single quotes * Typesetting \filename with \textsf instead of \textit * Remove forgotten \emph inside nonnormative * Say 'stronger requirement than being balanced' * Remove LaTeX source comment replaced by #2510 * Don't write math-like equation as Modelica source listing * Use \emph instead of \lstinline for 'platform' in context of LibraryDirectory * Wrap 'enabled' in \emph instead of \lstinline * Add double quotes around fragment of inline code when explaining multiple iterators * Use '\textgreater{} 2' instead of '$> 2$' * Write ="Hz" inside \lstinline instead of just 'Hz' * Write \lstinline!unit="s"! instead of 'unit \emph{s}' * Fix quoting of opening parenthesis character * Use \filename instead of \lstinline for some modelica URIs * Fix encoding of backslashes inside \filename * Fix matching of single quotes * Say \lstinline!'+'! instead of just \lstinline!+! when speaking of operator overloading * Say 'positive edge of' instead of 'when ... becomes ...' * Wrap content of \filename in \mbox to prevent hyphenation from injecting hyphens
Clarify that locally non-singular can combine to give a global singularity. (#2529)
Note that locally non-singular may combine to globally non-singular or globally singular; and the latter is the exception that should be mentioned. Closes #2510
Extend variability rules and fix invalid example (#2526)
* Fix and extend example on variability rules * Clarify the role of rules based on expression variability * Elaborate a little bit on the 'perfect matching rule' * A variable declared with 'constant' prefix is unaffected by the initialization problem Closes #2525
The upload of the files failed due to some symbolic links being broken. Also note that the upload script reads symbolic links instead of uploading them as they are.
Move nonnormative content regarding 'revisions' to separate paragraph
This is just the first of many similar changes to come. This commit mainly serves to populate the cleanup/nonnormative branch with some changes so that a pull request can be initiated.
Note that even this first change fixes a language syntax issue in addition that was missed due to the nonnormative content being in the middle of a paragraph.
New environment for non-normative content without delimiting paragraph breaks
When using the nonnormative* environment, it is the author's responsibility to ensure that the content doesn't mix with other paragraphs. The main use at the moment is in tables, where there shouldn't be a paragraph break before the line break at the end of a table row.
Comments set in italics are now formatted as non-normative content. The few remaining comments that are not formatted as non-normative content have no special formatting at all.
Tentatively remove all mathescape=true used with \lstinline
This is just to make the HTML build to work while investigating the problem with mathescape=true. Obviously, if we don't find a solution, just removing mathescape=true won't be enough; we'll also need change the actual math parts into something else, and then we should take the opportunity to do this consistently.