textcomp makes \textbullet smaller when using helvet font

Question

I'm writing a document which requires the helvet font and, in order to get access to the euro symbol, I am now trying to load the textcomp package.

However this package has the strange side effect of making the symbol produced by \textbullet, e.g. as used by an itemize environment, a bit smaller. I could live with that, but I sort of prefer the slightly larger bullet, so I'm wondering if there is an easy way to load textcomp but without changing the size of the bullet?

A small example showing the problem

\documentclass{article}
\usepackage[T1]{fontenc}
\usepackage[scaled]{helvet}
\renewcommand{\familydefault}{\sfdefault}
% Uncomment the following line to see the problem 
%\usepackage{textcomp}

\begin{document}
\begin{itemize}
\item One
\item Two
\end{itemize}
\end{document}

Answer 1

First, if you really want the origin \textbullet:

\DeclareTextSymbolDefault{\textbullet}{OMS}

---

Second, which is better? I know it is subjective, but I just want to explain what textcomp do.

With textcomp package, the text symbols are redeclared in new TS1 encoding, instead of Knuth's OMS encoding (designed for origin CM fonts). It is suitable for 8 bits fonts (256 glyphs).

If you use OMS encoding for \textbullet, then \textbullet is the same as

{\usefont{OMS}{phv}{m}{n}\symbol{15}}

And check omsphv.fd you'll find

\DeclareFontShape{OMS}{phv}{m}{n}
   {<-> ssub * cmsy/m/n}{}

i.e. Computer Modern Symbol font (cmsy) is used. In fact, it is not very suitable for Adobe Helvetica font (phv family).

If textcomp is used, \textbullet is in TS1 encoding, the result is the same as

{\usefont{TS1}{phv}{m}{n}\symbol{136}}

And check ts1phv.fd you'll find

\DeclareFontShape{TS1}{phv}{m}{n}{
   <-> \Hv@@scale phvr8c
}{}

That's real Helvetica. (Actully Nimbus Sans in most TeX distributions)

Answer 2

Don't use textcomp for the Euro symbol. There's eurosym, which provides the official symbol.

If you want to input the Euro symbol directly with the Unicode character and the utf8 option to inputenc, you can write

\usepackage{eurosym,newunicodechar}
\newunicodechar{€}{\officialeuro}

You won't be able to find € in the PDF, though. For that you need to add some magic:

\immediate\pdfobj stream {
  /CIDInit /ProcSet findresource begin
  12 dict begin
  begincmap
  /CIDSystemInfo
  << /Registry (TeX)
  /Ordering (Euro)
  /Supplement 0
  >> def
  /CMapName /TeX-Euro-0 def
  /CMapType 2 def
  1 begincodespacerange
  <00> <FF>
  endcodespacerange
  1 beginbfchar
  <65> <20AC>
  endbfchar
  endcmap
  CMapName currentdict /CMap defineresource pop
  end
  end
}
\begingroup
\edef\next#1#2{%
  \noexpand\fontseries{#1}\noexpand\fontshape{#2}\noexpand\selectfont
  \pdffontattr\font{/ToUnicode \the\pdflastobj\space 0 R}}
\fontencoding{U}\fontfamily{eurosym}%
\next{m}{n}\next{m}{sl}\next{m}{ol}
\next{bx}{n}\next{bx}{sl}\next{bx}{sl}
\endgroup

Basically this adds a CMAP resource and maps all eurosym fonts to it.