This repository contains the LaTeX templates for the journal Texto Livre and the final articles after the peer review and editorial process.

The file textolivre.cls implements the LaTeX class to create PDFs and the file textolivre-html.cls implements the LaTeX class to create HTMLs. To create a new article using the template, use the file article.tex as a starting point and read the instructions it provides. The template files should not be edited. If you require an additional package, insert it directly in the header of your .tex file.

creating your article.tex

Copy the example file article.tex and use it as your starting point. Note that the .tex file must be encoded in UTF-8. Use the journal document class textolivre:

\documentclass{textolivre}

The class provides the template for portuguese, english, spanish and french. The default behaviour is to select portuguese as the manuscript main language. You may switch to a different language by specifying it as a parameter when loading the class.

\documentclass[english]{textolivre}

packages

The provided template already uses a few LaTeX packages. You might check which are already included by running:

cat template/textolivre.cls | sed -n -e 's/\(^\\RequirePackage\)\(\[[-,a-z0-9=]*\]\)\?{\(.*\)}/\3/p' | sort | tr '\n' ','

As a result, we see that the packages already included are: abstract, adjustbox, amsmath, amsfonts, amssymb, amsthm, authblk, biblatex, caption, subcaption, cleveref, cmbright, csquotes, datetime2, enumitem, etoolbox, xpatch, fancyhdr, footmisc, geometry, graphicx, hyperref, hyperref, listings, longtable, booktabs, tabularx, colortbl, mfirstuc, polyglossia, relsize, setspace, textpos, titlesec, totpages, translations, xcolor, xstring.

Any other additional package might be included in the preamble of your .tex file.

language support

Multilanguage support is included through the polyglossia package. The main language should be setted when loading the class. The class will also set a secondary (and tertiary language, if necessary). Other languages might be used by means of the \setotherlanguage{...} command. The text in language, other than the default one, should come inside this language environment. Read the instruction in the template usage example provided. For further information on the polyglossia package, visit https://ctan.org/pkg/polyglossia.

title, author, abstract, keywords

Provided the title of the document using the standard \title command and the title in other language using the command \othertitle defined in the template. You might use \othertitle as many times as desired (usually just once and three times seldom).

Authors are inserted using the authblk package. For more information on the package, visit https://www.ctan.org/pkg/authblk. You might also provide the author ORCID by the \orcid command (see the usage in the example file).

To create multiple abstracts use the environment polyabstracts. For the abstract using the default language, just provide it using the usual abstract environment. For an abstract in another language, use the language environment provided by polyglossia. You may use as many abstracts as you need. See the example:

\begin{polyabstract}
\begin{abstract}
...

\keywords{word1 \sep word2 ...}
\end{abstract}

\begin{french}
\begin{abstract}
...

\keywords{mot1 \sep mot2 ...}
\end{abstract}
\end{french}
\end{polyabstract}

The example above also shows the usage of keywords. Use \sep between words.

compiling your document

Use XeLaTeX to compile you document using the command:

xelatex article.tex

If a PDF/A is required, compile using:

xelatex -shell-escape -output-driver="xdvipdfmx -z 0" article.tex

note that we have opted (-z 0) to suppress compression in XMP Metadata packet, in order to create a PDF/A compliant.

-shell-escape is required by pdfx. See the package documentation: https://ctan.org/pkg/pdfx. If you don't feel safe using it, you need to comment the PDF/A block in the end of the template file (textolivre.cls).

bibliography

Use biber to generate a bibliography in LaTeX. Biber has full Unicode support, that means your .bib might have any unicode character.

biber article

creating a PDF/A compliant file

Fill in the metadata information in your .tex file. See the example file. You will also need to install the package icc-profiles and convert all images to PDF/A files. To convert a PDF figure into a PDF/A version, use the following command in gv:

gs -dPDFA -dBATCH -dNOPAUSE -sColorConversionStrategy=UseDeviceIndependentColor -dCompatibilityLevel=1.4 -sDEVICE=pdfwrite -sProcessColorModel=DeviceCMYK -dPDFACompatibilityPolicy=2 -sOutputFile=figure-a.pdf figure.pdf

If you are using a raster image format, such as PNG, convert it to EPS and then to PDF/A:

convert figure.png figure.eps
gs -dPDFA -dBATCH -dNOPAUSE -sColorConversionStrategy=UseDeviceIndependentColor -dCompatibilityLevel=1.4 -sDEVICE=pdfwrite -sProcessColorModel=DeviceCMYK -dPDFACompatibilityPolicy=2 -dEPSCrop -sOutputFile=figure.pdf figure.eps

To batch convert multiple files, use the following script:

for img in $( ls figure*.png ); do convert $img ${img%png}eps; gs -dPDFA -dBATCH -dNOPAUSE -sColorConversionStrategy=UseDeviceIndependentColor -dCompatibilityLevel=1.4 -sDEVICE=pdfwrite -sProcessColorModel=DeviceCMYK -dPDFACompatibilityPolicy=2 -dEPSCrop -sOutputFile=${img%png}pdf ${img%png}eps; done

compiling errors

Some errors might occur when changing from one class into another, since the compiler tries to use intermediary files previously generated. To overcome these errors, just remove those auxiliary files. Usually removing the .aux will do. If you wish to remove all of them, here is the command:

find . -name 'article.*' ! -name '*.tex' ! -name '*.pdf' ! -name '*.bib' ! -name '*.html' -type f -exec rm {} \;

compiling into HTML

Compile the document using make4ht.

It might be necessary to get the updated .4ht files at: https://github.com/michal-h21/files/blob/master/tex4ht-4ht-files.zip

We also use some hooks written in the configuration file textolivre-html.cfg. To compile, use the command

make4ht -e build.lua -c textolivre.cfg -x -u article "fn-in,svg,pic-align"

we have added the parameter fn-in to have inline footnotes; svg for dvi pictures in svg format and pic-align for pictorial align environment. If you wish to get MathML equations, use

make4ht -e build.lua -c textolivre.cfg -u -x article "fn-in,mathml,mathjax"

French manuscrits also require the file gloss-french.4ht available at the template folder. See https://tex.stackexchange.com/questions/580456/error-when-using-polyglossia-french-and-make4ht

The HTML generated by make4ht has many breaklines which have been misinterpreted by some tools (like Marcalyc). If you need to remove them and any other cleaning/prettifying, use tidy. The following command might be used to remove breaklines

tidy -m --output-xhtml --break-before-br --wrap 0 article.html 2> errs.txt

For more information on tidy, visit https://www.html-tidy.org/documentation/.

creating the .zip file for Marcalyc

Although the HTML created by make4ht with the -u,--utf8 encodes the output document in utf8 and even writes the tag <meta charset='utf-8' />, Markalyc only accepts if it find the tag <meta content="text/html; charset=UTF-8" http-equiv="content-type" />. Use sed to change this tag:

sed -i 's/<meta\ charset=\x27utf-8\x27\ \/>/<meta\ content="text\/html;\ charset=UTF-8"\ http-equiv="content-type"\ \/>/' article.html

You should have created the HTMLs as described above. Now you need to zip all files and change the article.html name into index.html. The following command will create this intended zip using the current folder name as the name of the zip file:

ARTICLE=${PWD##*/} && zip ${ARTICLE}.zip article.html *.svg *.css *.png *.jpg && printf "@ article.html\n@=index.html\n" | zipnote -w ${ARTICLE}.zip