- ntg-context - Mailing Lists

Hobby's Graph package
by David Arnold 05 Sep '05

05 Sep '05

All, I find myself working my way through John Hobby's "Drawing Graphs with Metapost" again. In Figure 9, the following code has me confused: glabel.lft(image(unfill bbox lab[j]; draw lab[j]), length p[j]); I understand that the glabel command will put the label in this case at the "knot" length p[j]. But when he adds .7, I don't understand why the labels move down. glabel.lft(image(unfill bbox lab[j]; draw lab[j]), 0.7+length p[j]); Can someone help me with this?

2 1

RE: [NTG-context] another dvipdfmx/context bug
by Idris Samawi Hamid 05 Sep '05

05 Sep '05

>> ================================================ Dear Hans, It appears that there is a bug in the ctxtools. Consider original-public-lm.map. Lines such as cmbsy10 cmbsy10 cmbsy5 cmbsy5 cmbsy6 cmbsy7 should be either cmbsy10 default cmbsy10 cmbsy5 default cmbsy5 cmbsy6 default cmbsy7 partucularly when the name of the tfm and pfb do differ; or just cmbsy10 cmbsy5 cmbsy6 where the name of the tfm and pfb do not differ. Is this right ChoF? Best Idris >> about.dvi -> about.pdf >> [1 >> ** ERROR ** Could not find encoding file "cmsy10". >> >> Output file removed. >> ================================================ >> >> Please advise-) > >You may have an incorrect mapping file, original-public-lm.map. >I guess that the mapping file contains the following line > >cmsy10 cmsy10 > >which is incorrect. Just change as follows: > >cmsy10 > >It's enough. It seems that ctxtools does not generate a right mapping >file. ============================ Professor Idris Samawi Hamid Department of Philosophy Colorado State University Fort Collins, CO 80523

1 0

another dvipdfmx/context bug
by Idris Samawi Hamid 05 Sep '05

05 Sep '05

Dear gang, Here is anotehr dvipdfmx bug. I already did ctxtools --dpxmapfiles --force texmf ctxtools --dpxmapfiles --force texmf-local ctxtools --dpxmapfiles --force texmf-fonts Now the following file fails: ================================================ % output=dvipdfmx \starttext \startitemize \item test \stopitemize \stoptext ================================================ dvipdfmx fails with ================================================ about.dvi -> about.pdf [1 ** ERROR ** Could not find encoding file "cmsy10". Output file removed. ================================================ Please advise-) Best Idris ============================ Professor Idris Samawi Hamid Department of Philosophy Colorado State University Fort Collins, CO 80523

2 1

Antykwa math support & \Longrightarrow
by Mojca Miklavec 05 Sep '05

05 Sep '05

Hello, a slightly strange question. I re-compiled a pretty old ConTeXt document a couple of days ago. In the "old good times" it was typeset with \setupbodyfont[ant,18pt] and an equation with $a\Longrightarrow b$. Apparently math support was added inbetween. I'm not asking for any backward compatibility at all (my system behaves pretty strange anyway), I would only like to know the most elegant way to keep Antykwa for "a" and "b" in the example above, but to use a latin-modern arrow. I'm sorry, I wasn't able to reproduce the result anywhere else (each of the three distributions where I can try out ConTeXt is behaving differently), so I temporary uploded the files to http://pub.mojca.org/tex/ant/. And the second question: Why does this code below fail? It wotks ok for texnansi, but not for ec: \usetypescript[antykwa-torunska][ec] \definetypeface[MyAnt] [rm] [serif] [antykwa-torunska] [default] [encoding=ec] \setupbodyfont[MyAnt,10pt] \starttext abc\ccaron \stoptext Thank you very much, Mojca

2 2

Some concerns about the ConTeXt license.
by Brooks Moses 05 Sep '05

05 Sep '05

I have a number of concerns about wording of the ConTeXt license (as given in the readme file), which I'd like to bring up. In virtually all cases, my concern is simply that the actual wording of the license does not appear to be sufficient to provide the permissions that it intends to -- that is, the letter of the license does not live up to its spirit. (Note that I am not a lawyer, and that none of the below comments should be construed as legal advice.) I'm going by the text on http://wiki.contextgarden.net/Read_Me, which I believe to be the current version. ---- First, starting with section 3.1: "ConTeXt is available as free software in the spirit of the gnu general public licence." "In the spirit of the GPL" is not really useful in a license -- my idea of what's "in the spirit of" may be vastly different from Hans's idea of it. So this unfortunately offers nothing solid with regards to permission. It is, at best, a statement of intent. So, what else is there in the license that actually and clearly offers me a grant of permission to redistribute? I can find only the following: "2.7: ConTeXt is available for free for everyone who want to use it. Distributors are free to choose between the latest official release and the (in most cases stable) beta version." "7.2: We hereby grant permission to the de facto official TeX distributors teTeX, fpTeX and TeXlive to replace this licence with their own, i.e. their licence may supersede or enhance this one." "7.3: ConTeXt may be distributed as part of a public distribution in the spirit of free distributions." In addition, there is the following comment, which appears to be a restriction, but is relevant here: "3.3: The official version (i.e. the original zipped archive recognized by the prefix cont-) may only be redistributed as a whole and under this licence." To analyze those in turn: 2.7 does not offer any permission to redistribute; all it appears to offer is that ConTeXt will not charge for downloads. It implicitly offers "distributors" the opportunity to distribute the latest official release and the beta version, but it does not offer permission for any other versions. Of the items in section 7, 7.2 is not useful to the end user, as teTeX seem to incorporate the existing license rather than superceding it. Section 7.3 is, again, merely a description of intent with "in the spirit of" language rather than a clear grant of permission. Finally, 3.3 appears to restrict all distributions of any part of ConTeXt, as there is no "unofficial" version available from Pragma, and there is no provision for anyone else creating one -- we can only redistribute the "official" version as a whole, and thus it remains the official version. Thus, in sum: there is what appears to be an implicit grant for people to distribute the latest release or the current beta version, so long as it is complete and unchanged. There is nothing that I see as a clear grant of permission to distribute an unofficial modified copy, nor of permission to extract a portion of the code for use in another project, except for the "in the spirit of the GPL" language, which is IMHO very insufficient for this. Further, there is not even a clear grant of permission for the distribution of non-current official versions, aside again from the "in the spirit of" language. ---- Suppose that we consider the above issues to be addressed, by treating section 3.1 as saying effectively "The permissions granted under the GPL are also granted by this license." Let us now consider how the rest of the license actually agrees with that (and with section 4.1: "This sections will not introduce limitations in use that conflict with the gnu general public licence."). First, section 2.7, as quoted above. The GPL would allow distribution of any released version, whether current or not. It is not clear whether 2.7 is supposed to be a restriction of this or merely a statement about what Pragma offers. Section 4.2: "However, the interface of such modules should fit the general scheme philosophy." In the context of a file which purports to be a license file, this appears to be a legal requirement, not merely as a friendly guideline. If treated as a requirement, this would not be in the spirit of the GPL. Section 4.3: "Keywords and command names should conform to the already defined ones and therefore not clash with existing functionality." Again, same problem as 4.2. Section 4.5: "This also means that for instance in the texmf directory structure, the context/base, context/extra and context/sample locations are reserved for the official distribution." This prohibits the distribution of a modified unofficial version, which is contrary to the spirit of the GPL, though perhaps in the spirit of the LPPL. It also appears to restrict what I can personally do on my own computer, which is definitely contrary to the spirit of the GPL. Section 4.8: "When adapted versions of ConTeXt are distributed, these versions must contain the original distribution as a separate item." This prohibits me from taking, say, thirty lines of ConTeXt code and using them in a LaTeX package I'm writing (with, of course, appropriate citation), unless I distribute the entirety of ConTeXt with the package. This is very much contrary to the spirit of the GPL. Section 4.8, again: "The runtime versions (format files) must have different names...." See all the discussions on the Debian mailing list about whether this is in the spirit of Free Software. The latest version of the LPPL backs off slightly on the "no using the same filename" restriction, for this reason. It is also unclear whether the "\mycntxt\base\cont-xxx.tex" is a "different name" from \context\base\cont-xxx.tex" or not; I think this should be addressed. Section 4.8: again: "... and changed files must have an additional \writestatus message, expressing that something is changed." This, at present, would seem to apply to anything that uses any ConTeXt code, even if it is very obviously not part of ConTeXt (e.g., the hypothetical LaTeX package mentioned earlier). If taken as a general restriction, this is not in the spirit of the GPL, in my opinion; the restriction should apply only to things that could otherwise be mistaken for an official distribution. Section 4.9: "Those who write extensions or change files, must make sure that the original ConTeXt does not suffer from loading the wrong files." This is a restriction on parts of my system that may not be derived from ConTeXt at all, which is definitely against the spirit of the GPL. It is fair to place this as a restriction on "systems that include an Official distribution of ConTeXt", however. Section 4.11: "The functionality however must not be changed." Again, this is contrary to the spirit of the GPL, unless it is restricted only to things that are described as official releases of ConTeXt. ---- Some miscellaneous comments on things that are otherwise unclear. Section 4.7: "Distributers may add configuration settings to cont-usr and cont-sys." It's not clear who qualifies as a "distributor" in this statement. Section 4.11: "The auxiliary programs may be adapted to local conditions, such as operating systems specifics. The functionality however must not be changed." It is not clear here what is an auxiliary program. Section 7.4: "Each distribution of ConTeXt should include a copy of this text." Is a "distribution" the entirety of ConTeXt, or is it anything that includes any ConTeXt code? Does this mean that I need to include the whole text about Pragma's ConTeXt philosophy with anything that uses any ConTeXt code? ---- Some final notes: The redistribution requirements include the requirement that the text of the README file/license be included in the distributed version. This is incompatible with section 3.2, which says "This is the official version". For that reason, and a number of others, I think that the license and the readme should be separated -- and, in particular, there should be a clear delineation between what are Pragma's statements about the official release (including any promises of support), what are restrictions for redistributing ConTeXt as an "official release", what is part of the redistribution-as-free-software license, and what are friendly but non-binding suggestions for making modifications in the spirit of ConTeXt. My particular recommendation would be that the ConTeXt license should explicitly offer me the following three options for redistribution, in order to live up to its intent of being "in the spirit of free software": 1.) Redistribution of the entirety as an "official version of ConTeXt". This requires that all of the compatibility and completeness requirements be met. If the redistribution is part of a system, the entire system must meet the compatibility requirements -- e.g., no overriding ConTeXt functionality improperly. Redistribution must use the same license as the official version, with exceptions for teTeX et al. 2.) Redistribution as an "unofficial version based on ConTeXt". This permits redistribution of the system as a whole with modifications, but requires that the distribution clearly distinguish itself from the official version, and that it not call itself "ConTeXt" or be presented as such. (This may also require inclusion of an official version as well as the unofficial version.) The unofficial version must not install itself on the /context directories of a TeX distribution, all changed files must explicitly declare themselves to be changed, and so forth. Modifications must include all source code, when appropriate. These redistributions (including modifications) must use the same license as the original, again. 3.) Redistribution and reuse of portions of ConTeXt in other projects. This permits reuse of up to (for example) 30% of the ConTeXt code in other projects, without the requirements of distributing a full version. These portions can be taken as licensed either under the GPL or the LPPL, with no additional requirements other than provided by those licenses. I would be quite willing to elaborate on these suggestions, if requested. - Brooks

1 0

Cells table with long text
by Jilani Khaldi 05 Sep '05

05 Sep '05

Hi All, In a table I have some cells with long text (say 100 chars). How can I fit the text in a cell having many rows instead of a single line, as it is done (automatically) in HTML tables? An example will be very helpful. Thank you. jk -- // Jilani KHALDI http://www.archsf.org

2 4

Good link for stats teachers
by David Arnold 05 Sep '05

05 Sep '05

All, If you teach stats and you like Metapost, consider: http://www-math.univ-poitiers.fr/~phan/statsmac.html

1 0

A TABLE bug (?) from the Wiki.
by Brooks Moses 05 Sep '05

05 Sep '05

The Wiki page on TABLE contains the following example: ---- \setupTABLE[r][each][height=1cm] \setupTABLE[c][each][width=1cm] \setupTABLE[r][1][height=0cm,frame=off] \bTABLE \bTR \bTD \eTD \bTD \eTD \bTD \eTD \bTD \eTD \bTD \eTD \bTD \eTD \eTR \bTR \bTD[nc=2,nr=2] r1c1 \eTD \bTD r1c3 \eTD \bTD[nr=2] r1c4 \eTD \bTD[nr=2,nc=2] r1c5 \eTD \eTR \bTR \bTD[nr=2] r2c3 \eTD \eTR \bTR \bTD[nc=2] r3c1 \eTD \bTD[nc=2] r3c4 \eTD \bTD r3c6 \eTD \eTR \bTR \bTD r4c1 \eTD \bTD[nc=2] r4c2 \eTD \bTD[nr=2] r4c4 \eTD \bTD[nc=2] r4c5 \eTD \eTR \bTR \bTD[nr=2,nc=2] r5c1 \eTD \bTD[nr=2] r5c3 \eTD \bTD[nr=2,nc=2] r5c5 \eTD \eTR \bTR \bTD r6c4 \eTD \eTR \eTABLE ---- This comes with a note: "There is an ugly hack in the code. An empty first row with zero height and no borders is added. If the row is omitted, then something odd happens to the number of columns." Is this an actual bug in ConTeXt, or is it a bug in the example? If the latter, what's the fix? Thanks, - Brooks

2 1

How this LaTeX code could be translated into ConTeXt?
by Jilani Khaldi 05 Sep '05

05 Sep '05

Hi All, Migrating to Context, I need to translate this code from Latex \begin{table} \begin{tabularx}{16cm}{|c|X|X|} \hline blabla & blabla & blabla\\ \hline ... \hline \end{tabularx} \end{table} Any hint? Thanks! -- // Jilani KHALDI http://www.archsf.org

5 11

A number of footnote bugs.
by Brooks Moses 04 Sep '05

04 Sep '05

In playing with footnotes on the Wiki, I've come across a number of bugs, as well as the questions I mentioned earlier. Consider the following example text: \setupfootnotes[n=3] \starttext \strut\vfill % A hack to shorten the page, for Wiki use. This\footnote[footA](Or that\footnote{Or the other.}, if you prefer.} is a sentence with a footnote\footnote{Actually, two footnotes; this one and \in{footnote}[footA] on \at{page}[footA], denoted by \note[footA].}. \stoptext This illustrates the following bugs: 1.) Without the \setupfootnotes line, this leaves a vertical space of about one blank line between the bottom of the text and the footnote separator. With the \setupfootnotes line, we see that this space is proportional to n -- for n=3, it leaves three blank lines. 2.) The columns are ill-spaced in the default case; the footnote markers are closer to the column on their left than they are to the column they belong to. 3.) Footnotes in footnotes do not produce their numbers in the correct order, though they do set the footnotes in the proper order on the bottom of the page. Specifically, the second footnote in the text is (incorrectly) numbered 3 rather than 2, but it is (correctly) placed before the footnote on the footnote. 4.) According to the comments on lines 63-65 of core-not.tex, setting n=0 should set the footnotes in a continuous paragraph, using an algorithm from the TeXbook. This does not work at all; the footnotes are set in a single column but with a smaller interline space. Hans, are any of these readily fixable? - Brooks

1 0