2

I am facing an annoying problem with LaTeX which has occurred just recently. LaTeX refuses to compile and gives me the following error message:

"Runaway argument? File ended while scanning use of \abs@aux@page."

(Its probably not the at-sign, it just looks kinda like it.)

Now, the point is, there is nothing wrong with the document. I can run LaTeX, it works perfectly, and then I go to the kitchen for five minutes and come back and then I get this message when I try to compile. Meanwhile I have not changed a single thing. Nor does it matter what changes I do to the document, the error message will remain.

I have to go to my LaTeX folder and delete all of the auxiliary files apart from the .tex file itself, and then it works again. Without any changes. But then it comes back again. Particularly, if I actually do something wrong, anything wrong, and get another error message. Once I remove the error, it switches automatically to the error message above, and I have to go delete the files. Today I have deleted the auxiliary files close to fifty times.

Here is an example. I could make it more minimal, but then perhaps it's relevant to know all the packages loaded, since the error could be anywhere...

\documentclass{article}
\usepackage[english]{babel}
\usepackage[utf8]{inputenc}
\usepackage{csquotes}
\usepackage{authblk}
\usepackage [T1]{fontenc}
\usepackage[backend=biber, style=authoryear-comp]{biblatex} 
\usepackage{xyling}
\usepackage[super]{nth}
\usepackage{linguex}
\usepackage{xspace}
\usepackage{cgloss4e}
\usepackage{abstract}
\renewcommand*{\eachwordone}{\itshape}
\usepackage{booktabs}
\addbibresource{ref.bib}

\begin{document}

\renewcommand{\absnamepos}{empty}{The syntactial phenomenon known as verb-second or V2 is above all associated with the Germanic languages, whereas the Romance family is generally non-V2 or SVO-languages. However, it is common in literature to analyze all the major Old Romance languages as following a V2 pattern at some point in their historical development. In this paper, I argue that the origins of Medieval Romance V2 can be sought in certain pragmatic configurations of Spoken Latin, and I seek to corroborate this hypotohesis through a sample corpus from four Vulgar Latin texts.}

...etc.

Stefan Pinnow
  • 29,535
EspenJK
  • 519
  • 1
    without an example document probably not much help anyone can offer. – David Carlisle Mar 15 '16 at 18:40
  • applying (somewhat cracked) crystal ball ... is it possible that you have used \abstract{...} rather than \begin{abstract} ... \end{abstract}? how abstracts are defined depends intimately on what document class you are using, but the environment structure is the most common, and if it's expected, then if \end{abstract} isn't found, the job will never complete properly. – barbara beeton Mar 15 '16 at 18:40
  • 2
    make the smallest document that you can that shows the error, add \tracingall and post the document and the resulting log. – David Carlisle Mar 15 '16 at 18:48
  • If the minimal example is as easy as you claim it is, then that is a reason to post one, not to not post one. I can say with a large degree of certainty that I have never had that error and I've constructed many hundreds of documents (if not more) by now. So all your documents must be doing something that not everyone does --- and an MWE will quickly reveal what that is. – jon Mar 15 '16 at 19:11
  • being curious, i looked for instances of aux@page in the library of latex packages. in biblatex.sty there is a command \abx@aux@page, which seems like a possible misspelling of what you say is your problem. are you using biblatex perhaps? (really, an example document is needed!) – barbara beeton Mar 15 '16 at 20:33
  • Ok, I have added an example of the preamble and the start of the document. @barbarabeeton, I am indeed using biblatex, so perhaps the mistake could be related to that. But it's strange, because everything has been working fine, and now all of a suddenly not. And why is everything fine when I delete the auxiliary files? And if some error turns up, ANY error, then it will "convert" into this error once I correct what is wrong in the syntax. Suggestions appreciated... Thanks. – EspenJK Mar 16 '16 at 15:19
  • Does your example ( + \end{document}) reproduce the error for you? – samcarter_is_at_topanswers.xyz Mar 16 '16 at 16:20
  • 1
    Exemplum nimis minimum: The error (for me) would need to come in the 'etc.' part you haven't included. Note also that \renewcommand{\absnamepos}{empty} is the real command: the following part ({The syntactial ... Vulgar Latin texts.}) is not part of the redefinition: it is only a bunch of text that you have grouped through the use of braces. – jon Mar 16 '16 at 16:39
  • ... and looking at the docs (and code) for the abstract package, you should only redefine \absnamepos to an existing environment. The way you have it now, the absract package puts the abstract name in an\begin{empty} ...\end{empty}, which just so happens not to through an error. The normal choices for the redefinition are: center, flushleft, and flushright. – jon Mar 16 '16 at 16:50
  • The problem is that the "error" comes and goes. So, @samcarter, this example might or might not reproduce the error for me. It could compile, and I could go the store, and then it refuses to compile again at the very same spot when I come back. And, as I have said, if there is another error, any error, let's say I have made a mistake in a table which prompts a syntax error; I correct the mistake, and then the error almost invariably "converts" into this error. It must be related to the aux-files and the "re-reading", probably relate to biblatex. If I deleted the auxiliary files, it works... – EspenJK Mar 16 '16 at 16:55
  • Thanks, @jon, for pointing that out, I will fix that. But the problem seems rather to be related to biblatex. As Barbara pointed out, there is command called \abx@aux@page in the biblatex.sty... – EspenJK Mar 16 '16 at 17:03
  • Well, is the error abx or abs? Those are not the same at all. If abx, then we know it is biblatex. And please construct a file that will cause the error: even if it 'sometimes' causes the error, the file must be one that does so. I'm guessing the error creeps in after you run latex and biber and then on the next latex run the error kicks in. But we need the file, not something 'like' the file (which is why I wrote that your file is too minimal). – jon Mar 16 '16 at 17:12
  • Also, if it is abx@aux@page, please edit the question title and content so it clearly reflects the error. These questions and answers are meant to be helpful (in theory) to others down the road. – jon Mar 16 '16 at 17:13
  • It is indeed abx, I have edited the post accordingly. But as I have tried to explain, it is not possible to create a document that causes the error or at least I don't know how to do that. I can only give you the preamble and parts of the document WHEN it actually occurs. Should I try to attach some auxiliary file or something? – EspenJK Mar 16 '16 at 18:11
  • 1
    The procedure for stripping down the document (and it is more tedious when the .aux file is involved), is to slowly comment out the parts of the document between \begin{document} and \end{document}. If you look carefully at the text surrounding the error message, you should be able to figure out approximately where the problem arises. 1. Make an exact copy of your document and use that for testing. 2. Comment out the other text, delete the auxiliary files and see if you can recreate the error. ... (con't) – jon Mar 16 '16 at 18:51
  • 1
    ... 3. Then if it seems biblatex related (seems likely), try commenting out the non-biblatex-related packages and see if you can still recreate the error. It is tedious work, certainly, but if you want help we need a minimal document to look at. – jon Mar 16 '16 at 18:52
  • the fact that the error comes and goes -- and never appears if you don't have ,aux files -- indicates that it is indeed something generated during the "first processing" and only read back in during a subsequent run. since you have verified the problem command as \abx@aux@page, look into your .aux and other "derived" files to see if that command is contained therein. if it's not, then it may be necessary to try to identify the command that triggered it. in biblatex.sty, all instances of the command are associated with page tracking. – barbara beeton Mar 16 '16 at 19:05
  • Thanks to both of you for your helpful comments and suggestions. I will try to follow your suggestions, and then get back here if I find out something useful. – EspenJK Mar 16 '16 at 21:31

0 Answers0