<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"> <html> <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"/> <head> <title>Penlight Documentation</title> <link rel="stylesheet" href="../ldoc_fixed.css" type="text/css" /> </head> <body> <div id="container"> <div id="product"> <div id="product_logo"></div> <div id="product_name"><big><b></b></big></div> <div id="product_description"></div> </div> <!-- id="product" --> <div id="main"> <!-- Menu --> <div id="navigation"> <br/> <h1>Penlight</h1> <ul> <li><a href="../index.html">Index</a></li> </ul> <h2>Contents</h2> <ul> <li><a href="#Working_with_Paths">Working with Paths </a></li> <li><a href="#File_Operations">File Operations </a></li> <li><a href="#Directory_Operations">Directory Operations </a></li> </ul> <h2>Manual</h2> <ul class="$(kind=='Topics' and '' or 'nowrap'"> <li><a href="../manual/01-introduction.md.html">Introduction</a></li> <li><a href="../manual/02-arrays.md.html">Tables and Arrays</a></li> <li><a href="../manual/03-strings.md.html">Strings. Higher-level operations on strings.</a></li> <li><strong>Paths and Directories</strong></li> <li><a href="../manual/05-dates.md.html">Date and Time</a></li> <li><a href="../manual/06-data.md.html">Data</a></li> <li><a href="../manual/07-functional.md.html">Functional Programming</a></li> <li><a href="../manual/08-additional.md.html">Additional Libraries</a></li> <li><a href="../manual/09-discussion.md.html">Technical Choices</a></li> </ul> <h2>Libraries</h2> <ul class="$(kind=='Topics' and '' or 'nowrap'"> <li><a href="../libraries/pl.html">pl</a></li> <li><a href="../libraries/pl.Set.html">pl.Set</a></li> <li><a href="../libraries/pl.app.html">pl.app</a></li> <li><a href="../libraries/pl.array2d.html">pl.array2d</a></li> <li><a href="../libraries/pl.class.html">pl.class</a></li> <li><a href="../libraries/pl.compat.html">pl.compat</a></li> <li><a href="../libraries/pl.comprehension.html">pl.comprehension</a></li> <li><a href="../libraries/pl.config.html">pl.config</a></li> <li><a href="../libraries/pl.data.html">pl.data</a></li> <li><a href="../libraries/pl.dir.html">pl.dir</a></li> <li><a href="../libraries/pl.file.html">pl.file</a></li> <li><a href="../libraries/pl.func.html">pl.func</a></li> <li><a href="../libraries/pl.import_into.html">pl.import_into</a></li> <li><a href="../libraries/pl.input.html">pl.input</a></li> <li><a href="../libraries/pl.lapp.html">pl.lapp</a></li> <li><a href="../libraries/pl.lexer.html">pl.lexer</a></li> <li><a href="../libraries/pl.luabalanced.html">pl.luabalanced</a></li> <li><a href="../libraries/pl.operator.html">pl.operator</a></li> <li><a href="../libraries/pl.path.html">pl.path</a></li> <li><a href="../libraries/pl.permute.html">pl.permute</a></li> <li><a href="../libraries/pl.pretty.html">pl.pretty</a></li> <li><a href="../libraries/pl.seq.html">pl.seq</a></li> <li><a href="../libraries/pl.sip.html">pl.sip</a></li> <li><a href="../libraries/pl.strict.html">pl.strict</a></li> <li><a href="../libraries/pl.stringio.html">pl.stringio</a></li> <li><a href="../libraries/pl.stringx.html">pl.stringx</a></li> <li><a href="../libraries/pl.tablex.html">pl.tablex</a></li> <li><a href="../libraries/pl.template.html">pl.template</a></li> <li><a href="../libraries/pl.test.html">pl.test</a></li> <li><a href="../libraries/pl.text.html">pl.text</a></li> <li><a href="../libraries/pl.types.html">pl.types</a></li> <li><a href="../libraries/pl.url.html">pl.url</a></li> <li><a href="../libraries/pl.utils.html">pl.utils</a></li> <li><a href="../libraries/pl.xml.html">pl.xml</a></li> </ul> <h2>Classes</h2> <ul class="$(kind=='Topics' and '' or 'nowrap'"> <li><a href="../classes/pl.Date.html">pl.Date</a></li> <li><a href="../classes/pl.List.html">pl.List</a></li> <li><a href="../classes/pl.Map.html">pl.Map</a></li> <li><a href="../classes/pl.MultiMap.html">pl.MultiMap</a></li> <li><a href="../classes/pl.OrderedMap.html">pl.OrderedMap</a></li> </ul> <h2>Examples</h2> <ul class="$(kind=='Topics' and '' or 'nowrap'"> <li><a href="../examples/seesubst.lua.html">seesubst.lua</a></li> <li><a href="../examples/sipscan.lua.html">sipscan.lua</a></li> <li><a href="../examples/symbols.lua.html">symbols.lua</a></li> <li><a href="../examples/test-cmp.lua.html">test-cmp.lua</a></li> <li><a href="../examples/test-data.lua.html">test-data.lua</a></li> <li><a href="../examples/test-listcallbacks.lua.html">test-listcallbacks.lua</a></li> <li><a href="../examples/test-pretty.lua.html">test-pretty.lua</a></li> <li><a href="../examples/test-symbols.lua.html">test-symbols.lua</a></li> <li><a href="../examples/testapp.lua.html">testapp.lua</a></li> <li><a href="../examples/testclone.lua.html">testclone.lua</a></li> <li><a href="../examples/testconfig.lua.html">testconfig.lua</a></li> <li><a href="../examples/testglobal.lua.html">testglobal.lua</a></li> <li><a href="../examples/testinputfields.lua.html">testinputfields.lua</a></li> <li><a href="../examples/testinputfields2.lua.html">testinputfields2.lua</a></li> <li><a href="../examples/testxml.lua.html">testxml.lua</a></li> <li><a href="../examples/which.lua.html">which.lua</a></li> </ul> </div> <div id="content"> <h2>Paths and Directories</h2> <p><a name="Working_with_Paths"></a></p> <h3>Working with Paths</h3> <p>Programs should not depend on quirks of your operating system. They will be harder to read, and need to be ported for other systems. The worst of course is hardcoding paths like 'c:\' in programs, and wondering why Vista complains so much. But even something like <code>dir..'\'..file</code> is a problem, since Unix can't understand backslashes in this way. <code>dir..'/'..file</code> is <em>usually</em> portable, but it's best to put this all into a simple function, <a href="../libraries/pl.path.html#join">path.join</a>. If you consistently use <a href="../libraries/pl.path.html#join">path.join</a>, then it's much easier to write cross-platform code, since it handles the directory separator for you.</p> <p><a href="../libraries/pl.path.html#">pl.path</a> provides the same functionality as Python's <code>os.path</code> module (11.1).</p> <pre> > p = <span class="string">'c:\\bonzo\\DOG.txt'</span> > = path.normcase (p) <span class="comment">---> only makes sense on Windows </span>c:\bonzo\dog.txt > = path.splitext (p) c:\bonzo\DOG .txt > = path.extension (p) .txt > = path.basename (p) DOG.txt > = path.exists(p) <span class="keyword">false</span> > = path.join (<span class="string">'fred'</span>,<span class="string">'alice.txt'</span>) fred\alice.txt > = path.exists <span class="string">'pretty.lua'</span> <span class="keyword">true</span> > = path.getsize <span class="string">'pretty.lua'</span> <span class="number">2125</span> > = path.isfile <span class="string">'pretty.lua'</span> <span class="keyword">true</span> > = path.isdir <span class="string">'pretty.lua'</span> <span class="keyword">false</span> </pre> <p>It is very important for all programmers, not just on Unix, to only write to where they are allowed to write. <a href="../libraries/pl.path.html#expanduser">path.expanduser</a> will expand '~' (tilde) into the home directory. Depending on your OS, this will be a guaranteed place where you can create files:</p> <pre> > = path.expanduser <span class="string">'~/mydata.txt'</span> <span class="string">'C:\Documents and Settings\SJDonova/mydata.txt'</span> > = path.expanduser <span class="string">'~/mydata.txt'</span> /home/sdonovan/mydata.txt </pre> <p>Under Windows, <a href="http://www.lua.org/manual/5.2/manual.html#pdf-os.tmpname">os.tmpname</a> returns a path which leads to your drive root full of temporary files. (And increasingly, you do not have access to this root folder.) This is corrected by <a href="../libraries/pl.path.html#tmpname">path.tmpname</a>, which uses the environment variable TMP:</p> <pre> > <span class="global">os</span>.tmpname() <span class="comment">-- not a good place to put temporary files! </span><span class="string">'\s25g.'</span> > path.tmpname() <span class="string">'C:\DOCUME~1\SJDonova\LOCALS~1\Temp\s25g.1'</span> </pre> <p>A useful extra function is <a href="../libraries/pl.path.html#package_path">pl.path.package_path</a>, which will tell you the path of a particular Lua module. So on my system, <code>package_path('pl.path')</code> returns 'C:\Program Files\Lua\5.1\lualibs\pl\path.lua', and <code>package_path('ifs')</code> returns 'C:\Program Files\Lua\5.1\clibs\lfs.dll'. It is implemented in terms of <a href="http://www.lua.org/manual/5.2/manual.html#pdf-package.searchpath">package.searchpath</a>, which is a new function in Lua 5.2 which has been implemented for Lua 5.1 in Penlight.</p> <p><a name="File_Operations"></a></p> <h3>File Operations</h3> <p><a href="../libraries/pl.file.html#">pl.file</a> is a new module that provides more sensible names for common file operations. For instance, <a href="http://www.lua.org/manual/5.2/manual.html#pdf-file:read">file.read</a> and <a href="http://www.lua.org/manual/5.2/manual.html#pdf-file:write">file.write</a> are aliases for <a href="../libraries/pl.utils.html#readfile">utils.readfile</a> and <a href="../libraries/pl.utils.html#writefile">utils.writefile</a>.</p> <p>Smaller files can be efficiently read and written in one operation. <a href="http://www.lua.org/manual/5.2/manual.html#pdf-file:read">file.read</a> is passed a filename and returns the contents as a string, if successful; if not, then it returns <code>nil</code> and the actual error message. There is an optional boolean parameter if you want the file to be read in binary mode (this makes no difference on Unix but remains important with Windows.)</p> <p>In previous versions of Penlight, <a href="../libraries/pl.utils.html#readfile">utils.readfile</a> would read standard input if the file was not specified, but this can lead to nasty bugs; use <code>io.read '*a'</code> to grab all of standard input.</p> <p>Similarly, <a href="http://www.lua.org/manual/5.2/manual.html#pdf-file:write">file.write</a> takes a filename and a string which will be written to that file.</p> <p>For example, this little script converts a file into upper case:</p> <pre> <span class="global">require</span> <span class="string">'pl'</span> <span class="global">assert</span>(#arg == <span class="number">2</span>, <span class="string">'supply two filenames'</span>) text = <span class="global">assert</span>(file.read(arg[<span class="number">1</span>])) <span class="global">assert</span>(file.write(arg[<span class="number">2</span>],text:upper())) </pre> <p>Copying files is suprisingly tricky. <a href="../libraries/pl.file.html#copy">file.copy</a> and <a href="../libraries/pl.file.html#move">file.move</a> attempt to use the best implementation possible. On Windows, they link to the API functions <code>CopyFile</code> and <code>MoveFile</code>, but only if the <code>alien</code> package is installed (this is true for Lua for Windows.) Otherwise, the system copy command is used. This can be ugly when writing Windows GUI applications, because of the dreaded flashing black-box problem with launching processes.</p> <p><a name="Directory_Operations"></a></p> <h3>Directory Operations</h3> <p><a href="../libraries/pl.dir.html#">pl.dir</a> provides some useful functions for working with directories. <code>fnmatch</code> will match a filename against a shell pattern, and <code>filter</code> will return any files in the supplied list which match the given pattern, which correspond to the functions in the Python <code>fnmatch</code> module. <code>getdirectories</code> will return all directories contained in a directory, and <code>getfiles</code> will return all files in a directory which match a shell pattern. These functions return the files as a table, unlike <a href="http://stevedonovan.github.io/lua-stdlibs/modules/lfs.html#dir">lfs.dir</a> which returns an iterator.)</p> <p><a href="../libraries/pl.dir.html#makepath">dir.makepath</a> can create a full path, creating subdirectories as necessary; <code>rmtree</code> is the Nuclear Option of file deleting functions, since it will recursively clear out and delete all directories found begining at a path (there is a similar function with this name in the Python <code>shutils</code> module.)</p> <pre> > = dir.makepath <span class="string">'t\\temp\\bonzo'</span> > = path.isdir <span class="string">'t\\temp\\bonzo'</span> <span class="keyword">true</span> > = dir.rmtree <span class="string">'t'</span> </pre> <p><a href="../libraries/pl.dir.html#rmtree">dir.rmtree</a> depends on <a href="../libraries/pl.dir.html#walk">dir.walk</a>, which is a powerful tool for scanning a whole directory tree. Here is the implementation of <a href="../libraries/pl.dir.html#rmtree">dir.rmtree</a>:</p> <pre> <span class="comment">--- remove a whole directory tree. </span><span class="comment">-- @param path A directory path </span><span class="keyword">function</span> dir.rmtree(fullpath) <span class="keyword">for</span> root,dirs,files <span class="keyword">in</span> dir.walk(fullpath) <span class="keyword">do</span> <span class="keyword">for</span> i,f <span class="keyword">in</span> <span class="global">ipairs</span>(files) <span class="keyword">do</span> <span class="global">os</span>.remove(path.join(root,f)) <span class="keyword">end</span> lfs.rmdir(root) <span class="keyword">end</span> <span class="keyword">end</span> </pre> <p><a href="../libraries/pl.dir.html#clonetree">dir.clonetree</a> clones directory trees. The first argument is a path that must exist, and the second path is the path to be cloned. (Note that this path cannot be <em>inside</em> the first path, since this leads to madness.) By default, it will then just recreate the directory structure. You can in addition provide a function, which will be applied for all files found.</p> <pre> <span class="comment">-- make a copy of my libs folder </span><span class="global">require</span> <span class="string">'pl'</span> p1 = <span class="string">[[d:\dev\lua\libs]]</span> p2 = <span class="string">[[D:\dev\lua\libs\..\tests]]</span> dir.clonetree(p1,p2,dir.copyfile) </pre> <p>A more sophisticated version, which only copies files which have been modified:</p> <pre> <span class="comment">-- p1 and p2 as before, or from arg[1] and arg[2] </span>dir.clonetree(p1,p2,<span class="keyword">function</span>(f1,f2) <span class="keyword">local</span> res <span class="keyword">local</span> t1,t2 = path.getmtime(f1),path.getmtime(f2) <span class="comment">-- f2 might not exist, so be careful about t2 </span> <span class="keyword">if</span> <span class="keyword">not</span> t2 <span class="keyword">or</span> t1 > t2 <span class="keyword">then</span> res = dir.copyfile(f1,f2) <span class="keyword">end</span> <span class="keyword">return</span> res <span class="comment">-- indicates successful operation </span><span class="keyword">end</span>) </pre> <p><a href="../libraries/pl.dir.html#clonetree">dir.clonetree</a> uses <a href="../libraries/pl.path.html#common_prefix">path.common_prefix</a>. With <code>p1</code> and <code>p2</code> defined above, the common path is 'd:\dev\lua'. So 'd:\dev\lua\libs\testfunc.lua' is copied to 'd:\dev\lua\test\testfunc.lua', etc.</p> <p>If you need to find the common path of list of files, then <a href="../libraries/pl.tablex.html#reduce">tablex.reduce</a> will do the job:</p> <pre> > p3 = <span class="string">[[d:\dev]]</span> > = tablex.reduce(path.common_prefix,{p1,p2,p3}) <span class="string">'d:\dev'</span> </pre> </div> <!-- id="content" --> </div> <!-- id="main" --> <div id="about"> <i>generated by <a href="http://github.com/stevedonovan/LDoc">LDoc 1.4.3</a></i> <i style="float:right;">Last updated 2015-03-21 18:49:03 </i> </div> <!-- id="about" --> </div> <!-- id="container" --> </body> </html>