use the new hlabel command to properly create links in PDFs

fixes #44

.github/workflows/compile.yml

@@ -26,6 +26,9 @@ jobs:
     - name: Install pre-requisites for spell check
       working-directory: master
       run: ./dev/install-spellcheck.sh
+    - name: Check labels
+      working-directory: master
+      run: ./dev/check-label.sh
     - name: Run spellcheck
       working-directory: master
       run: make spellcheck

common.tex

@@ -69,6 +69,10 @@
 \renewcommand{\slidelabel}{}
 \setlength{\textwidth}{0.9\textwidth}
 
+% This command has to be used throught the sources instead of \label
+% for hyperref to be able to create valid links.
+\newcommand{\hlabel}{\phantomsection\label}
+
 \newcommand{\sltitle}[1]{{\centering\textbf{\Large #1}
     \vskip 2em plus 0pt minus 2em\par}} % Slide title
 

dev/check-label.sh

@@ -0,0 +1,10 @@
+#!/bin/bash
+
+if grep '\\label{.*}' *.tex >/tmp/label-err.out; then
+	echo "Some of the *.tex files contain the \\label command."
+	echo "This is unwanted as this leads to invalid links in PDFs."
+	echo "Use \\hlabel instead."
+	echo ""
+	cat /tmp/label-err.out
+	exit 1
+fi

file-api.tex

@@ -54,7 +54,7 @@
 \end{center}
 \end{slide}
 
-\label{OPENFILETABLES}
+\hlabel{OPENFILETABLES}
 
 \begin{itemize}
 \item This is a simplified view of the kernel tables that deal with files.  It
@@ -113,7 +113,7 @@
 \end{itemize}
 \end{slide}
 
-\label{OPEN}
+\hlabel{OPEN}
 
 \begin{itemize}
 \item The first available file descriptor is always used.
@@ -145,7 +145,7 @@
 reading.
 \item \texttt{O\_NOCTTY} \dots{} when opening a terminal by a process without a
 controlling terminal, the terminal being opened does not become one.
-\item \label{O_NONBLOCK} \texttt{O\_NONBLOCK} \dots{} if reading or writing
+\item \hlabel{O_NONBLOCK} \texttt{O\_NONBLOCK} \dots{} if reading or writing
 cannot be satisfied right away, calls \texttt{read}/\texttt{write} will fail
 instead of getting blocked and waiting for the completion.  \texttt{errno} is
 set to \texttt{EAGAIN} in such a case.
@@ -187,8 +187,8 @@
 \end{itemize}
 \end{slide}
 
-\label{MKFIFO}
-\label{CREAT}
+\hlabel{MKFIFO}
+\hlabel{CREAT}
 
 \begin{itemize}
 \item The \texttt{open} call allows opening of a regular file, device, or named
@@ -230,7 +230,7 @@
 \end{itemize}     
 \end{slide}
 
-\label{READCALL}
+\hlabel{READCALL}
 
 \begin{itemize}
 \setlength{\itemsep}{0.8\itemsep}
@@ -306,7 +306,7 @@
 and it is a long living process, a daemon perhaps, depending on the system
 configuration you may hit memory starvation as the system memory will be filled
 with an ever growing file descriptor table.
-\item \label{SIMPLE_CAT} A very simple \texttt{cat(1)} program:
+\item \hlabel{SIMPLE_CAT} A very simple \texttt{cat(1)} program:
 \example{read/cat.c}
 \end{itemize}
 
@@ -386,7 +386,7 @@
 \end{itemize}
 \end{slide}
 
-\label{NAMEDPIPE}
+\hlabel{NAMEDPIPE}
 
 \begin{itemize}
 \item A named pipe is created using the system call \texttt{mkfifo}, see page
@@ -491,7 +491,7 @@
 \end{slide}
 
 \begin{itemize}
-\item \label{LSEEK} The first byte is at position 0.  If it makes sense, you may
+\item \hlabel{LSEEK} The first byte is at position 0.  If it makes sense, you may
 use a negative number for setting the \emph{offset}.  Example:
 \example{read/lseek.c}.
 \item It is legal to move beyond the end of the file.  If data is written there,
@@ -505,7 +505,7 @@
 \item You can obviously use the return value of \texttt{lseek} not only for
 subsequent calls to \texttt{read} and \texttt{write} but also for another call
 to \texttt{lseek}.
-\item \label{BIG_FILE} Beware of files with holes as it may lead to problems
+\item \hlabel{BIG_FILE} Beware of files with holes as it may lead to problems
 with backing up the data.  Example: \example{read/big-file.c} demonstrates that
 moving a sparse file may end up in an actual storage data occupation increase.
 It greatly depends on the system you run, what archiving utility is used, and
@@ -568,7 +568,7 @@
 \end{itemize}
 \end{slide}
 
-\label{DUP_CALL}
+\hlabel{DUP_CALL}
 
 \begin{itemize}
 \item We already know that the first available file descriptor is used when
@@ -610,7 +610,7 @@
 \begin{itemize}
 \item Note the flag \texttt{O\_APPEND} used to implement a redirection
 \texttt{>>}.
-\item \label{REDIRECT} Another example of \texttt{dup} use will be provided when
+\item \hlabel{REDIRECT} Another example of \texttt{dup} use will be provided when
 we start working with pipes.  The first redirection example from the slide
 (without \texttt{stderr}) is in \example{read/redirect.c}.  In that example, the
 \texttt{execl} call replaces the current process image with the
@@ -682,7 +682,7 @@
 \end{itemize}
 \end{slide}
 
-\label{FCNTL}
+\hlabel{FCNTL}
 
 \begin{itemize}
 \item Example: close standard error output on program execution (exec call):\\
@@ -790,7 +790,7 @@
 last modification of the inode.
 \item The UNIX norm does not specify the ordering of the \texttt{struct stat}
 members, nor does it prohibit adding new ones.
-\item \label{STAT} Example: \example{stat/stat.c}
+\item \hlabel{STAT} Example: \example{stat/stat.c}
 \item You can call \texttt{fstat} on file descriptors 0, 1, and 2 as well.  Unless
 redirected before, you will get information on the underlying terminal device
 (e.g. \texttt{/dev/ttys011} on macOS).  Example: \example{stat/stat012.c}.
@@ -897,7 +897,7 @@
 \emph{path2} to file \emph{path1}.  Hard links cannot span filesystems (use
 \funnm{symlink} for that).
 \end{itemize}
-\label{UNLINK} \texttt{ int \funnm{unlink}(const char *\emph{path});}
+\hlabel{UNLINK} \texttt{ int \funnm{unlink}(const char *\emph{path});}
 \begin{itemize}
 \item deletes a name (i.e. a directory entry) and after deleting the last link to
 the file and after closing the file by all processes, deletes the file data.
@@ -1028,7 +1028,7 @@
 mountpoint; this member contains the i-node number of the directory where the
 filesystem is mounted and not the root of the mounted filesystem as one might
 expect.
-\item \label{D_TYPE} Some system have a \texttt{struct dirent} member
+\item \hlabel{D_TYPE} Some system have a \texttt{struct dirent} member
 \texttt{d\_type}.  It can have values of \texttt{DT\_REG}, \texttt{DT\_DIR},
 \texttt{DT\_FIFO} etc., see the man page for \texttt{dirent}.  It was a BSD
 specific thing and subsequently copied by other systems, including Linux.
@@ -1038,7 +1038,7 @@
 to remove all entries before you can delete a directory.  You can use
 \texttt{system("rm -r xxx")} etc. but be careful to sanitize the environment and
 the directory name to avoid any security implications.
-\item \label{REMOVE} The specification also defines \texttt{remove} which
+\item \hlabel{REMOVE} The specification also defines \texttt{remove} which
 behaves like \texttt{unlink} for files and as \texttt{rmdir} for directories.
 \end{itemize}
 

files.tex

@@ -30,7 +30,7 @@
 \end{itemize}
 \end{slide}
 
-\label{DEVFS}
+\hlabel{DEVFS}
 
 \begin{itemize}
 \item Devices, files in \texttt{/proc}, terminals, memory etc. are of one type
@@ -433,7 +433,7 @@
 
 %%%%%
 
-\label{VFS}
+\hlabel{VFS}
 \pdfbookmark[1]{Virtual File System}{VFS}
 
 \begin{slide}

intro.tex

@@ -169,7 +169,7 @@
 \end{itemize}
 \end{slide}
 
-\label{UNIXSTANDARDS}
+\hlabel{UNIXSTANDARDS}
 
 \begin{itemize}
 \item The very basic information is that the area of UNIX standards is very
@@ -244,7 +244,7 @@
 \end{itemize}
 \end{slide}
 
-\label{POSIX}
+\hlabel{POSIX}
 
 \begin{itemize}
 \item The first document is \emph{IEEE Std POSIX1003.1-1988}, formerly simply
@@ -346,7 +346,7 @@
 \end{enumerate}
 \end{slide}
 
-\label{REF_PROGRAMMING}
+\hlabel{REF_PROGRAMMING}
 
 \begin{enumerate}
 \item One of the best book on programming in Unix environment.  Does not cover
@@ -362,7 +362,7 @@
 recommended.
 \item UNIX specifications.
 \item Detailed descriptions of system calls and functions.
-\item \label{POSIX4} A book that did not fit the slide and covers topics outside
+\item \hlabel{POSIX4} A book that did not fit the slide and covers topics outside
 of the scope of this class: Gall\-meis\-ter, B. R.: \emsl{POSIX.4 Programmers
 Guide: Programming for the Real World}, O'Reilly; \nth{1} edition, 1995.  A great
 book on real-time POSIX extensions with a beautiful cover.  See also pages
@@ -410,7 +410,7 @@
 %%%%%
 
 \pdfbookmark[1]{The C Programming Language}{C}
-\label{C_LANGUAGE}
+\hlabel{C_LANGUAGE}
 
 \begin{slide}
 \sltitle{The C Programming Language}
@@ -499,7 +499,7 @@
 \end{itemize}
 \end{slide}
 
-\label{BYTE_ORDERING}
+\hlabel{BYTE_ORDERING}
 
 \begin{itemize}
 \item Be careful when using tools like \texttt{hexdump} that by default print
@@ -551,7 +551,7 @@
 \end{itemize}
 \end{slide}
 
-\label{NEWLINECHAR}
+\hlabel{NEWLINECHAR}
 
 \begin{itemize}
 \item \emsl{LF}, \emph{line feed}, sometimes also referred to as \emph{new
@@ -1092,7 +1092,7 @@
 \end{minipage}
 \end{slide}
 
-\label{MAKE}.
+\hlabel{MAKE}.
 
 \begin{itemize}
 \item You could compile and link the program via one invocation of a compiler,
@@ -1268,7 +1268,7 @@
 \end{itemize}
 \end{slide}
 
-\label{RUNTIMELINKER}
+\hlabel{RUNTIMELINKER}
 
 \begin{itemize}
 \item An ELF object format is explained on page \pageref{ELF}.
@@ -1365,7 +1365,7 @@
 \item You can also edit ELF objects via \texttt{elfedit(1)} on Solaris.  You can
 change \texttt{RUNPATH}, for example.
 \end{itemize}
-\item \label{EVIL_LDLIBPATH} In general, you should not use
+\item \hlabel{EVIL_LDLIBPATH} In general, you should not use
 \texttt{LD\_LIBRARY\_PATH} for anything else than debugging during the
 development or when moving libraries between directories.  You can find lots of
 articles on ``why is \texttt{LD\_LIBRARY\_PATH} evil?'' etc.  For example,
@@ -1413,7 +1413,7 @@
 \end{itemize}
 \end{slide}
 
-\label{API_ABI}
+\hlabel{API_ABI}
 
 \begin{itemize}
 \item In short -- an API is source code based while an ABI is binary based.
@@ -1482,7 +1482,7 @@
 -1077941135
 \end{verbatim}
 
-\item \label{ABI_MAIN} Example: \example{lib-abi/abi-main.c} (see the block
+\item \hlabel{ABI_MAIN} Example: \example{lib-abi/abi-main.c} (see the block
 comment in the file on how to use other files located in the same directory).
 \item To change an ABI safely, you need library versioning -- if the library
 ABI change is not backward compatible, a bumped up version needs to prevent
@@ -1867,7 +1867,7 @@
 different from an empty string.
 \item To go through all the command line arguments, you can either use
 \emph{argc} or test for \texttt{NULL} in \texttt{argv[i]}.
-\item \label{SHELL_ARGV0} \texttt{argv[0]} is sometimes a source of additional
+\item \hlabel{SHELL_ARGV0} \texttt{argv[0]} is sometimes a source of additional
 information.  For example, commands \texttt{cp}, \texttt{mv}, and \texttt{ln} may
 be linked to the same executable (Solaris).  The value of \texttt{argv[0]} then
 tells the process what function it is supposed to perform.  Another example -- if
@@ -1912,7 +1912,7 @@
 prog && echo "success" || echo "failure"
 \end{verbatim}
 Example: \example{main/return-256.c}.
-\item \label{RETURN255} Never use \texttt{return (-1)} in \texttt{main} nor
+\item \hlabel{RETURN255} Never use \texttt{return (-1)} in \texttt{main} nor
 \texttt{exit(-1)}.  Based on the information in the previous paragraph,
 from \texttt{-1} you will get \texttt{255} as the return value you get in the
 shell in \texttt{\$?}.  It just creates confusion.
@@ -1927,7 +1927,7 @@
 \emsl{without} printing a new line), and calls functions registered via
 \texttt{atexit()}, and possibly other actions based on a specific system.
 Example: \example{exit/exit.c} and \example{exit/atexit-abort.c}
-\item \label{MAIN_C} Example on printing out command line arguments:
+\item \hlabel{MAIN_C} Example on printing out command line arguments:
 \example{main/print-argv.c}
 \item If a process is killed by a signal, you can get the signal number from its
 return value as presented by the shell.  See page
@@ -2175,7 +2175,7 @@
 processing.
 \item When a undefined option is used, \texttt{getopt} will print an error;
 this can be suppressed by setting the \texttt{opterr} variable to 0.
-\item \label{GETOPT} Example: shell script \example{getopt/getopts.sh}
+\item \hlabel{GETOPT} Example: shell script \example{getopt/getopts.sh}
 rewritten to C language using the \texttt{getopt} function:
 \example{getopt/getopt.c}
 \end{itemize}
@@ -2499,7 +2499,7 @@
 \item Many \texttt{libc} functions and functions from other libraries set
 \texttt{errno} on failure as well.  You always need to consult the relevant
 documentation.
-\item \label{ERRNO} It is common nowadays that \texttt{errno} is in reality
+\item \hlabel{ERRNO} It is common nowadays that \texttt{errno} is in reality
 defined in \texttt{libc} as a dereferenced pointer to an integer returned by a
 function (returned pointer is specific to a userland thread) and the value is
 set right after the instruction for the system call.  For example, on the i386
@@ -2557,7 +2557,7 @@
 %%%%%
 
 \begin{itemize}
-\label{ERR}
+\hlabel{ERR}
 \item The \funnm{errx}() function behaves as \funnm{err}() but does not use
 \texttt{errno} to print that extra error message.  Similarly, \funnm{warnx}().
 \item These functions are very handy especially for smaller programs as they are