public void HTMLexport_fixedWordList()
{
// In this test, we use a small number of fixed items in the word
// list to get a stable output for regression testing.
//
// In particular, we use items that would fail to display correctly
// on a web page due to the wrong encoding (should be UTF-8).
//
Dictionary <string, string> someCaseCorrection =
new Dictionary <string, string>();
Dictionary <string, string> someWord2URL =
new Dictionary <string, string>();
//Dictionary<string, string> someCaseCorrection_Reverse =
// new Dictionary<string, string>();
// First
someCaseCorrection.Add("JS", "JavaScript");
someWord2URL.Add(
"JavaScript",
"https://en.wikipedia.org/wiki/JavaScript");
// Second
someCaseCorrection.Add("angstrom", "Ångström Linux");
someWord2URL.Add(
"Ångström Linux",
"https://en.wikipedia.org/wiki/%C3%85ngstr%C3%B6m_distribution");
// Third
someCaseCorrection.Add("utorrent", "µTorrent");
someWord2URL.Add(
"µTorrent", "http://en.wikipedia.org/wiki/%CE%9CTorrent");
string Wordlist_HTML =
WikipediaLookup.dumpWordList_asHTML(
"",
ref someCaseCorrection,
someWord2URL.Count,
ref someWord2URL
);
int len = Wordlist_HTML.Length;
// Poor man's hash: check length (later, use a real
// hashing function). At least it should catch that
// indentation in the HTML source is not broken
// by changes (e.g. refactoring) and unintended omissions
// deletions.
//
// But it will not detect single spaces replaced by single TAB...
//
Assert.AreEqual(
3572 - 24 + 153 +
36 + 85 + 4,
len, "XYZ");
// -24 because we removed unnecessary space...
// +153 because we added a justification for the existence
// of the word list...
// +36 because we changed the formatting of the CSS...
// +85 because we added a link to the web version
// of Edit Overflow...
// +4 Because we went to development mode again (after
// the release 2019-11-01)... But why +4 (the extra
// "a3" counts two times).
Assert.AreEqual(Wordlist_HTML.IndexOf("\t"), -1, "XYZ"); // Detect
// any TABs...
} //HTMLexport_fixedWordList()
static void Main(string[] args)
{
string outputType = Environment.GetEnvironmentVariable("WORDLIST_OUTPUTTYPE");
string toOutput = ""; // No output unless explicitly indicated
// by parameters passed to the program
EditorOverflowApplication app = new EditorOverflowApplication_Unix();
CodeFormattingCheck cfCheck = new CodeFormattingCheck();
// This will result in running the ***first*** level of
// integrity testing for the word list data
WikipediaLookup someWikipediaLookup = new WikipediaLookup();
switch (outputType)
{
case "SQL":
// That is what we most often use (to update the
// deployed database with the word list data).
//
// Like for HTML generation, it will result in
// running ***more*** rigorous integrity
// testing for the word list data.
toOutput = someWikipediaLookup.dumpWordList_asSQL();
break;
case "HTML":
// This will result in running ***more*** rigorous
// integrity testing for the word list data
//toOutput = someWikipediaLookup.dumpWordList_asHTML(
//
// // Fixed strings - sufficient for integrity testing
// // of the word list data
// //
// "some combined regular expressions",
// "some version thingie",
// "some date only string");
toOutput = someWikipediaLookup.dumpWordList_asHTML(
cfCheck.combinedAllOfRegularExpressions(),
app.fullVersionStr(),
app.versionString_dateOnly()
);
break;
default:
//Console.WriteLine("Hello, World!");
Console.WriteLine("\n");
Console.WriteLine("2020-02-28T012833Z+0");
Console.WriteLine("\n");
string var1 = Environment.GetEnvironmentVariable("PATH");
//Console.WriteLine("Environment variable PATH: " + var1 + "\n");
Console.WriteLine(
"Output type for wordlist not specified. " +
"Use environment variable WORDLIST_OUTPUTTYPE " +
"with 'SQL' or 'HTML'.\n");
break;
}
// Dump the SQL or HTML to standard output so we can
// redirect it to a file (but note that integrity
// error messages currently also end up there...).
//
// What about Unicode / UTF-8????????
Console.WriteLine(toOutput);
}
public void HTMLexport_emptyWordList()
{
// In this test, we call it with all empty variable information
// to get a stable output for regression testing (get skeleton
// HTML, with only the begining and end, with an empty
// word table)
Dictionary <string, string> someCaseCorrection =
new Dictionary <string, string>();
Dictionary <string, string> someWord2URL =
new Dictionary <string, string>();
Dictionary <string, string> someCaseCorrection_Reverse =
new Dictionary <string, string>();
string Wordlist_HTML =
WikipediaLookup.dumpWordList_asHTML(
"",
ref someCaseCorrection,
someCaseCorrection_Reverse.Count,
ref someWord2URL);
int len = Wordlist_HTML.Length;
// Poor man's hash: check length (later, use a real
// hashing function). At least it should catch that
// indentation in the HTML source is not broken
// by changes (e.g. refactoring) and unintended omissions
// deletions.
//
// But it will not detect single spaces replaced by single TAB...
//
Assert.AreEqual(
2708 + 3 + 1 + 12 - 10 + 1 + 8 - 22 + 9 - 2 - 1 - 1 + 1 +
404 + 153 +
36 + 85 + 4,
len, "XYZ");
// +3 because we discovered and eliminated a tab...
// +1 because changed the HTML slightly...
// +12 because we the end tag for "head" was missing...
// -10 because we used for proper formatting for a <p> tag...
// +1 because we made the HTML look more like in the browser...
// +8 because we fixed indentation for the HTML source for list items...
// -22 because we made the formatting for <p> tags consistent...
// +9 because we fixed the indentation for <hr/>...
// -2 because we made the formatting for the <table> tag consistent...
// -1 because we made the formatting for the table header
// line consistent...
// -1 because we made the formatting for the table end
// tag consistent...
// +1 because we made the formatting for <h2> and <p>
// tags consistent...
// +404 For a new baseline, after changes to the end of
// the HTML content.
// +153 because we added a justification for the existence
// of the word list...
// +36 because we changed the formatting of the CSS...
// +85 because we added a link to the web version
// of Edit Overflow...
// +4 Because we went to development mode again (after
// the release 2019-11-01)... But why +4 (the extra
// "a3" counts two times).
Assert.AreEqual(Wordlist_HTML.IndexOf("\t"), -1, "XYZ"); // Detect
// any TABs...
} //HTMLexport_emptyWordList()
