Dear All, Before writing these lines I tried to locate earlier discussions on tbl to html output, but given 1,390 hits on "tbl" in the search window alone, I did not find appropriate messages. I remember, though, that there are some very good reasons why tbl should be rendered as html and not as embedded graphics in html pages, especially if there are hyperlinks in the table pointing outwards, or if pointing into a huge table is desired. So, I wish I could find a working tbl -> html strategy for my problem.
My problem: I reconstructed approx. 1,400 tables with varying degrees of complexity and size; the two examples attached give a good indication of both ends of the spectrum. My status: All 1,400 tables were converted to pdf and png by wrapping the tbl code in a minimal .ms document the output of which (in case of pdf) was cropped to printed areas, or which was produced by the -Thtml driver. Everything looks as intended, so far. Now the publication goal shifts more and more towards html, I and think converting all tables to html should be my optimal goal. What I tried so far: - Eric Raymond's doclifter. I failed because I could not construct a minimal .ms document which would not be rejected by doclifter, and every time I try the error message relates to "internal errors". I fail at making doclifter accept documents as simple as the following one: .TL My Title .AU Oliver Corff .SH First Section .PP Text follows here. - I searched the net and found a roff package called troffcvt-1.04 which promises html output (http://www.snake.net/software/troffcvt/doc/troffcvt.html) which is by now more than 23 years old; I failed to compile the whole package on my modern system (and have been away from compiling my own sources for about the same time). The package contains a tbl preprocessor written in perl which produces an intermediate output format suitable for troffcvt's own processing pipeline. See the attached files tbl.tbl and tbl.cvtoutput. Possible options: - Attempt to convert troffcvt's tblcvt output to html by own perl script; the output format is well documented: http://www.snake.net/software/troffcvt/doc/tblcvt.html - Attempt to go via groff -Ttxt and convert that output to html; - Attempt to write my own restricted minimal tbl parser in perl which directly outputs html code. Given the number of input files, rewriting them by hand in html is not a good idea (if it were two handful, then ok). I have a good month or so which I could spend on tinkering a solution, but before I begin and run into dead ends, what would your suggestion be? Thank you very much, and please accept my apologies for this lengthy post. Oliver.
1979_0826_Politbuero_Mitglieder.pdf
Description: Adobe PDF document
1979_0843_Postleitzahlen.pdf
Description: Adobe PDF document
.TS tab(|); l l. One|Two Three|Four .TE
.T*table*begin 2 2 0 L n n n n .T*column*info 0 3 n .T*column*info 0 3 n .T*row*begin .T*cell*info L 1 1 C 0 .T*cell*info L 1 1 C 0 .T*cell*begin 0 0 0 O .T*cell*end .T*cell*begin 0 0 0 n .T*cell*end .T*row*end .T*row*begin .T*cell*info L 1 1 C 0 .T*cell*info L 1 1 C 0 .T*cell*begin 0 0 0 T .T*cell*end .T*cell*begin 0 0 0 h .T*cell*end .T*row*end .T*table*end
