index.html 8.8 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144
  1. <!doctype html>
  2. <title>CodeMirror: Q mode</title>
  3. <meta charset="utf-8"/>
  4. <link rel=stylesheet href="../../doc/docs.css">
  5. <link rel="stylesheet" href="../../lib/codemirror.css">
  6. <script src="../../lib/codemirror.js"></script>
  7. <script src="../../addon/edit/matchbrackets.js"></script>
  8. <script src="q.js"></script>
  9. <style type="text/css">.CodeMirror {border-top: 1px solid black; border-bottom: 1px solid black;}</style>
  10. <div id=nav>
  11. <a href="http://codemirror.net"><h1>CodeMirror</h1><img id=logo src="../../doc/logo.png"></a>
  12. <ul>
  13. <li><a href="../../index.html">Home</a>
  14. <li><a href="../../doc/manual.html">Manual</a>
  15. <li><a href="https://github.com/codemirror/codemirror">Code</a>
  16. </ul>
  17. <ul>
  18. <li><a href="../index.html">Language modes</a>
  19. <li><a class=active href="#">Q</a>
  20. </ul>
  21. </div>
  22. <article>
  23. <h2>Q mode</h2>
  24. <div><textarea id="code" name="code">
  25. / utilities to quickly load a csv file - for more exhaustive analysis of the csv contents see csvguess.q
  26. / 2009.09.20 - updated to match latest csvguess.q
  27. / .csv.colhdrs[file] - return a list of colhdrs from file
  28. / info:.csv.info[file] - return a table of information about the file
  29. / columns are:
  30. / c - column name; ci - column index; t - load type; mw - max width;
  31. / dchar - distinct characters in values; rule - rule that caught the type
  32. / maybe - needs checking, _could_ be say a date, but perhaps just a float?
  33. / .csv.info0[file;onlycols] - like .csv.info except that it only analyses <onlycols>
  34. / example:
  35. / info:.csv.info0[file;(.csv.colhdrs file)like"*price"]
  36. / info:.csv.infolike[file;"*price"]
  37. / show delete from info where t=" "
  38. / .csv.data[file;info] - use the info from .csv.info to read the data
  39. / .csv.data10[file;info] - like .csv.data but only returns the first 10 rows
  40. / bulkload[file;info] - bulk loads file into table DATA (which must be already defined :: DATA:() )
  41. / .csv.read[file]/read10[file] - for when you don't care about checking/tweaking the <info> before reading
  42. \d .csv
  43. DELIM:","
  44. ZAPHDRS:0b / lowercase and remove _ from colhdrs (junk characters are always removed)
  45. WIDTHHDR:25000 / number of characters read to get the header
  46. READLINES:222 / number of lines read and used to guess the types
  47. SYMMAXWIDTH:11 / character columns narrower than this are stored as symbols
  48. SYMMAXGR:10 / max symbol granularity% before we give up and keep as a * string
  49. FORCECHARWIDTH:30 / every field (of any type) with values this wide or more is forced to character "*"
  50. DISCARDEMPTY:0b / completely ignore empty columns if true else set them to "C"
  51. CHUNKSIZE:50000000 / used in fs2 (modified .Q.fs)
  52. k)nameltrim:{$[~@x;.z.s'x;~(*x)in aA:.Q.a,.Q.A;(+/&\~x in aA)_x;x]}
  53. k)fs2:{[f;s]((-7!s)>){[f;s;x]i:1+last@&0xa=r:1:(s;x;CHUNKSIZE);f@`\:i#r;x+i}[f;s]/0j}
  54. cleanhdrs:{{$[ZAPHDRS;lower x except"_";x]}x where x in DELIM,.Q.an}
  55. cancast:{nw:x$"";if[not x in"BXCS";nw:(min 0#;max 0#;::)@\:nw];$[not any nw in x$(11&count y)#y;$[11<count y;not any nw in x$y;1b];0b]}
  56. read:{[file]data[file;info[file]]}
  57. read10:{[file]data10[file;info[file]]}
  58. colhdrs:{[file]
  59. `$nameltrim DELIM vs cleanhdrs first read0(file;0;1+first where 0xa=read1(file;0;WIDTHHDR))}
  60. data:{[file;info]
  61. (exec c from info where not t=" ")xcol(exec t from info;enlist DELIM)0:file}
  62. data10:{[file;info]
  63. data[;info](file;0;1+last 11#where 0xa=read1(file;0;15*WIDTHHDR))}
  64. info0:{[file;onlycols]
  65. colhdrs:`$nameltrim DELIM vs cleanhdrs first head:read0(file;0;1+last where 0xa=read1(file;0;WIDTHHDR));
  66. loadfmts:(count colhdrs)#"S";if[count onlycols;loadfmts[where not colhdrs in onlycols]:"C"];
  67. breaks:where 0xa=read1(file;0;floor(10+READLINES)*WIDTHHDR%count head);
  68. nas:count as:colhdrs xcol(loadfmts;enlist DELIM)0:(file;0;1+last((1+READLINES)&count breaks)#breaks);
  69. info:([]c:key flip as;v:value flip as);as:();
  70. reserved:key`.q;reserved,:.Q.res;reserved,:`i;
  71. info:update res:c in reserved from info;
  72. info:update ci:i,t:"?",ipa:0b,mdot:0,mw:0,rule:0,gr:0,ndv:0,maybe:0b,empty:0b,j10:0b,j12:0b from info;
  73. info:update ci:`s#ci from info;
  74. if[count onlycols;info:update t:" ",rule:10 from info where not c in onlycols];
  75. info:update sdv:{string(distinct x)except`}peach v from info;
  76. info:update ndv:count each sdv from info;
  77. info:update gr:floor 0.5+100*ndv%nas,mw:{max count each x}peach sdv from info where 0<ndv;
  78. info:update t:"*",rule:20 from info where mw>.csv.FORCECHARWIDTH; / long values
  79. info:update t:"C "[.csv.DISCARDEMPTY],rule:30,empty:1b from info where t="?",mw=0; / empty columns
  80. info:update dchar:{asc distinct raze x}peach sdv from info where t="?";
  81. info:update mdot:{max sum each"."=x}peach sdv from info where t="?",{"."in x}each dchar;
  82. info:update t:"n",rule:40 from info where t="?",{any x in"0123456789"}each dchar; / vaguely numeric..
  83. info:update t:"I",rule:50,ipa:1b from info where t="n",mw within 7 15,mdot=3,{all x in".0123456789"}each dchar,.csv.cancast["I"]peach sdv; / ip-address
  84. info:update t:"J",rule:60 from info where t="n",mdot=0,{all x in"+-0123456789"}each dchar,.csv.cancast["J"]peach sdv;
  85. info:update t:"I",rule:70 from info where t="J",mw<12,.csv.cancast["I"]peach sdv;
  86. info:update t:"H",rule:80 from info where t="I",mw<7,.csv.cancast["H"]peach sdv;
  87. info:update t:"F",rule:90 from info where t="n",mdot<2,mw>1,.csv.cancast["F"]peach sdv;
  88. info:update t:"E",rule:100,maybe:1b from info where t="F",mw<9;
  89. info:update t:"M",rule:110,maybe:1b from info where t in"nIHEF",mdot<2,mw within 4 7,.csv.cancast["M"]peach sdv;
  90. info:update t:"D",rule:120,maybe:1b from info where t in"nI",mdot in 0 2,mw within 6 11,.csv.cancast["D"]peach sdv;
  91. info:update t:"V",rule:130,maybe:1b from info where t="I",mw in 5 6,7<count each dchar,{all x like"*[0-9][0-5][0-9][0-5][0-9]"}peach sdv,.csv.cancast["V"]peach sdv; / 235959 12345
  92. info:update t:"U",rule:140,maybe:1b from info where t="H",mw in 3 4,7<count each dchar,{all x like"*[0-9][0-5][0-9]"}peach sdv,.csv.cancast["U"]peach sdv; /2359
  93. info:update t:"U",rule:150,maybe:0b from info where t="n",mw in 4 5,mdot=0,{all x like"*[0-9]:[0-5][0-9]"}peach sdv,.csv.cancast["U"]peach sdv;
  94. info:update t:"T",rule:160,maybe:0b from info where t="n",mw within 7 12,mdot<2,{all x like"*[0-9]:[0-5][0-9]:[0-5][0-9]*"}peach sdv,.csv.cancast["T"]peach sdv;
  95. info:update t:"V",rule:170,maybe:0b from info where t="T",mw in 7 8,mdot=0,.csv.cancast["V"]peach sdv;
  96. info:update t:"T",rule:180,maybe:1b from info where t in"EF",mw within 7 10,mdot=1,{all x like"*[0-9][0-5][0-9][0-5][0-9].*"}peach sdv,.csv.cancast["T"]peach sdv;
  97. info:update t:"Z",rule:190,maybe:0b from info where t="n",mw within 11 24,mdot<4,.csv.cancast["Z"]peach sdv;
  98. info:update t:"P",rule:200,maybe:1b from info where t="n",mw within 12 29,mdot<4,{all x like"[12]*"}peach sdv,.csv.cancast["P"]peach sdv;
  99. info:update t:"N",rule:210,maybe:1b from info where t="n",mw within 3 28,mdot=1,.csv.cancast["N"]peach sdv;
  100. info:update t:"?",rule:220,maybe:0b from info where t="n"; / reset remaining maybe numeric
  101. info:update t:"C",rule:230,maybe:0b from info where t="?",mw=1; / char
  102. info:update t:"B",rule:240,maybe:0b from info where t in"HC",mw=1,mdot=0,{$[all x in"01tTfFyYnN";(any"0fFnN"in x)and any"1tTyY"in x;0b]}each dchar; / boolean
  103. info:update t:"B",rule:250,maybe:1b from info where t in"HC",mw=1,mdot=0,{all x in"01tTfFyYnN"}each dchar; / boolean
  104. info:update t:"X",rule:260,maybe:0b from info where t="?",mw=2,{$[all x in"0123456789abcdefABCDEF";(any .Q.n in x)and any"abcdefABCDEF"in x;0b]}each dchar; /hex
  105. info:update t:"S",rule:270,maybe:1b from info where t="?",mw<.csv.SYMMAXWIDTH,mw>1,gr<.csv.SYMMAXGR; / symbols (max width permitting)
  106. info:update t:"*",rule:280,maybe:0b from info where t="?"; / the rest as strings
  107. / flag those S/* columns which could be encoded to integers (.Q.j10/x10/j12/x12) to avoid symbols
  108. info:update j12:1b from info where t in"S*",mw<13,{all x in .Q.nA}each dchar;
  109. info:update j10:1b from info where t in"S*",mw<11,{all x in .Q.b6}each dchar;
  110. select c,ci,t,maybe,empty,res,j10,j12,ipa,mw,mdot,rule,gr,ndv,dchar from info}
  111. info:info0[;()] / by default don't restrict columns
  112. infolike:{[file;pattern] info0[file;{x where x like y}[lower colhdrs[file];pattern]]} / .csv.infolike[file;"*time"]
  113. \d .
  114. / DATA:()
  115. bulkload:{[file;info]
  116. if[not`DATA in system"v";'`DATA.not.defined];
  117. if[count DATA;'`DATA.not.empty];
  118. loadhdrs:exec c from info where not t=" ";loadfmts:exec t from info;
  119. .csv.fs2[{[file;loadhdrs;loadfmts] `DATA insert $[count DATA;flip loadhdrs!(loadfmts;.csv.DELIM)0:file;loadhdrs xcol(loadfmts;enlist .csv.DELIM)0:file]}[file;loadhdrs;loadfmts]];
  120. count DATA}
  121. @[.:;"\\l csvutil.custom.q";::]; / save your custom settings in csvutil.custom.q to override those set at the beginning of the file
  122. </textarea></div>
  123. <script>
  124. var editor = CodeMirror.fromTextArea(document.getElementById("code"), {
  125. lineNumbers: true,
  126. matchBrackets: true
  127. });
  128. </script>
  129. <p><strong>MIME type defined:</strong> <code>text/x-q</code>.</p>
  130. </article>