X-VM-v5-Data: ([nil nil nil nil nil nil nil t nil] [nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil]) Date: Wed, 18 Apr 90 16:00:31 CET From: Rainer Schoepf Organization: Inst. f. Theor. Physik d. Univ. Heidelberg Subject: Re: Aux file blues. To: LaTeX-L Mailing list In-Reply-To: Message of Wed, 18 Apr 90 14:33:06 MET from Status: R X-Status: X-Keywords: X-UID: 79 What we need is a general mechanism to store arbitrary information in the .aux file. Think of the following: - In list-like environments, one could remember the width of the longest label and adjust the list parameters for the following run(s). - In the amstex style option many of the alignments process their text text twice to determine the width of the columns, more specifically to determinewhere the equation tag should go. This is taken over from plain AMSTeX (nice wording, eh?) where Spivak did this to be compatible with plain TeX's \eqno primitive. With LaTeX we should devise a way to calculate things in the first run and use them in the second and subsequent runs. I have spent a bit of time on this problem and I think I have found a partial solution. For every information of this type there is a line of the form \extra@aux@info{}{}{} `originator' is needed to distinguish different types of information, the `sequence number' is to identify theplace where it is needed. But this is theeasy part. The real problem is to find a way to access this information in a efficient way. In a largeTeX environment one can easily implement \extra@aux@info to define a macro whose name is derived from the originator and the sequence number and which expands into the information. But this will not work for small systems. The only alternative I can think of at the moment is to read the .aux file (or perhaps a special extra info file) every time information is needed. Any ideas or comments? Rainer