I'm making my first attempt to typeset an Arabic poem in two columns using bidipoem in eledmac.

All is going well--except I cannot adjust the vertical space between the lines without messing everything up. Either the spacing is too compact, i.e. the diacritics in Arabic run together, or the line-numbering in eledmac goes haywire (and the Arabic text gets bunched up in strange ways.)

Is there a simple way to add a small amount of vertical space between lines in bidipoem traditionalpoem?

\documentclass[12pt, letterpaper]{article}
\XeTeXlinebreakskip = 0pt plus 5pt
\newfontfamily\arabicfont[Script=Arabic, Scale=1.5]{Amiri}
ماذَا يقُولُ الأديبُ فِيمَا & تَرْكِيبُهُ مُنْتَهَى الحُزُوم        
منْ سَبْعَةٍ جُمِعَتْ لديهِ & تُعَدُّ من صنعَةِ الحكيمِ

The real problems start to appear in the second to last line where the diacritics of the upper and lower lines are running into one another. Is there a way that I can separate the lines from one another and not mess up the line numbering and spacing?

minimal example![][1]

  • 1
    tex.stackexchange.com/q/126075/7883 is relevant. But your MWE does not compile: “Package xepersian Error: You have not specified any font for the main Persian text.” Also note that \defaultfontfeatures{…} should come before you set specific fonts, otherwise it won’t affect them all.
    – Thérèse
    Jun 7, 2015 at 4:52
  • And welcome to the site!
    – Thérèse
    Jun 7, 2015 at 4:53
  • I think the problem is the interaction between eledmac and the traditonalpoem. In general, eledmac has many problem with package which do their own typesseting. However, I know that @robert-alessi has made command in order to obtain such "traditional poem" in two column. I will ask you to send you an answer, before integrating it in eledmac.
    – Maïeul
    Jun 7, 2015 at 8:58
  • Thanks very much for this @Maïeul and Thérèse. I'm able to get something approximating proper line counting in traditional poem in eledmac by adjusting the \lineskip slightly, but if I go too far more than 3pts it throws off the line counting by a verse. Please do send me a solution. I'll be very grateful! :) Jun 7, 2015 at 13:59
  • I have asked to RObert, but, now, it quite slow to answer.
    – Maïeul
    Jun 7, 2015 at 17:32

1 Answer 1


One way to get out of this is to forget typesetting Arabic verse with bidipoem. As a provisional solution, I would suggest this (in which arabtex is used: adapt if you prefer unicode Arabic):

  1. put this in your preamble: % Arabic verse \setstanzaindents{0,0} \setcounter{stanzaindentsrepetition}{1} \newlength{\baytwidth} \setlength{\baytwidth}{.3\textwidth} \newcommand{\bayt}[2]{\centering% \makebox[\baytwidth][s]{\ta[fullvoc]{#2}}% {\hspace{20pt}}% \makebox[\baytwidth][s]{\ta[fullvoc]{#1}}}

    1. Then in your document, at the appropriate places (this is an example for the astanza environment; adapt for stanza): \begin{astanza} \bayt{yasqIkahA man banI 'l-`abAdi ra^sA}% {munta__|si__buN `I__|duhu 'il_A 'l-'a.ha__|di}\&[\footnotesize(\ta{al-munsari.hu})\hfill\null] \end{astanza}

\bayt{}{} takes two arguments: the first one is for the ṣadr, and the second is for the ʿaǧuz. You may even, if necessary, redefine baytwidth which is set by default to 0.3\textwidth.

The example above gives you this: enter image description here

Of course, you may insert any kind of critical note into each or both of the \bayt{}{} arguments.

  • Dear Robert, Thanks so much for this elegant solution. One question--my text switches back and forth from poetry to prose often. Do you have any elegant solutions for making sure that the prose formatting is not affected by the poetry? I.e. it seems to justify and center things in ways that are somewhat awkward at times. Jul 3, 2015 at 15:57
  • In which way are things wrongly justified? I'm very sorry for the delay, @m-pomerantz Will try to catch up on this. Jan 11, 2016 at 23:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.