diff options
Diffstat (limited to 'notes.dsc')
| -rw-r--r-- | notes.dsc | 69 |
1 files changed, 69 insertions, 0 deletions
diff --git a/notes.dsc b/notes.dsc new file mode 100644 index 0000000..2e68855 --- /dev/null +++ b/notes.dsc @@ -0,0 +1,69 @@ +Basser Lout Version 3 and the PostScript Document Structuring Conventions +Jeffrey H. Kingston +24 March 1994 + +The following notes detail all areas known to the author where Lout does +not fully conform to Version 3.0 of the PostScript Document Structuring +Conventions (DSC), as given in Appendix G of the PostScript Language +Reference Manual, second edition (PLRM). Page numbers refer to PLRM. +These are minor infringements which should not trouble anyone (I hope). +Please let me know of any problems. + +Language level and extensions. Lout does not produce %%LanguageLevel and +%%Extensions comments, implying that its output is Level 1 PostScript with +no extensions (p644). This is true except for PostScript which user-level +commands choose to add via @SetColour, @Graphic, @IncludeGraphic, and +@PrependGraphic commands. If this added PostScript is not Level 1 then +the absence of %%LanguageLevel and %%Extensions comments will be in +violation of the DSC. Lout's standard packages use only Level 1, which +should take care of most of the @SetColour, @Graphic, and @PrependGraphic +commands encountered in practice. If a %%LanguageLevel or %%Extensions +comment is encountered in an @IncludeGraphic file, Lout prints a warning +message but takes no other action. + +Color separation conventions (pp685-688). At present Lout generates none +of the comments described in this section. + +DocumentSuppliedResources. According to p671, each included document +(%%Begin(End)Document pair in PostScript, @IncludeGraphic file in Lout) +should have a corresponding "%%DocumentSuppliedResources: file" entry. +This contradicts p660, where each %%DocumentSuppliedResources entry is +said to correspond to a %%Begin(End)Resource pair. Lout does not record +%%Begin(End)Document pairs in the %%DocumentSuppliedResources list. + +EOF. According to p673, the %%EOF comment of an included document (i.e. +one enclosed in %%Begin(End)Document) will be ignored. However, p658 +does not mention this exception, and one Adobe PostScript interpreter at +least in the author's environment terminates at the first %%EOF whether +or not it lies inside %%Begin(End)Document. Accordingly, Lout strips +out any %%EOF comment within any included document during inclusion. + +Font inclusion. According to p678, "%%IncludeResource: font" comments +relating to commonly available fonts should be placed in the document +setup section. However no definition of which fonts fall into this +category is offered, so this requirement is not observed by Lout. +Instead, Lout generates one "%%IncludeResource: font" in the page setup +section for each font used on the corresponding page, except for fonts +used on the first page which are done once only in the document setup +section. This implies that fonts used on many pages but not the first +will generate many "%%IncludeResource: font" comments, one for each page +on which the font is used, which in turn implies that these fonts might +be downloaded many times (Lout itself never downloads any font, but a +document manager might react to an "%%IncludeResource: font" comment by +doing so). However, most fonts used by most people are already in the +printer, and in these cases "it is highly likely that the font server +or document manager would ignore the inclusion request, because the +font would already be available on the printer" (p678). + +PageResources. This comment is supposed to list all resources needed +or supplied on the current page except for procsets (p685). At present +Lout lists fonts but not encoding vectors under this comment. + +Definitions in EPS files. On p736, "it is strongly recommended that an +EPS file make its definitions in its own dictionary or dictionaries". +Lout -EPS does not follow this recommendation, since it is not clear where +to create this dictionary or dictionaries, given that the DSC states (p625) +that the prolog should be a sequence of procset definitions (after the +header section and defaults section) only. Perhaps each procset should be +in its own dictionary, as is done in lout/include/fig_prepend for example, +but this is not stated anywhere. |