print " div.prewrap {white-space: pre-wrap;}"
print "</style>"
print "<title>A Latin Macronizer</title>"
print "</head>"
print "<body>"
print '<h1><a href="' + scriptname + '">A Latin Macronizer</a></h1>'
print "<p>Please enter a Latin text! (A correctly macronized text can be entered, in which case the performance will be evaluated.)</p>"
# print '<p>Note: Input longer than 20000 characters will be truncated. (Sorry about that!)</p>'
print '<form action="' + scriptname + '" method="post">'
print '<p><textarea name="textcontent" cols="100" rows="20">'
if texttomacronize == "":
macronizedtext = ""
else:
try:
macronizer = Macronizer()
macronizer.settext(texttomacronize)
macronizedtext = macronizer.gettext(domacronize, alsomaius, performutov, performitoj, markambigs=False)
sys.stdout.write(macronizedtext)
except Exception as inst:
print inst.args[0]
macronizedtext = ""
print "</textarea><br>"
print '<input type="checkbox" name="macronize" value="on" %s> Mark long vowels. ' % (
"checked" if domacronize else ""
)
print u'<input type="checkbox" name="alsomaius" value="on" %s> Also mark <i>māius</i> etc.<br>' % (
"checked" if alsomaius else ""
)
print '<input type="checkbox" name="utov" value="on" %s> Convert u to v. ' % (
"checked" if performutov else ""
)
<title>A Latin Macronizer</title>
</head>
<body>""")
print('<h1><a href="' + scriptname + '">A Latin Macronizer</a></h1>')
print('<p>Please enter a Latin text!</p>')
# print('<p>Note: In order to avoid time out from the server, input longer than %s characters will be truncated. Sorry about that!</p>' % (TRUNCATETHRESHOLD))
print('<form action="' + scriptname + '" method="post">')
print(
'<p><textarea name="textcontent" onclick="enlarge(this)" cols="100" rows="%s">'
% ('20' if texttomacronize == "" else '3'))
if texttomacronize == "":
macronizedtext = ""
else:
try:
macronizer = Macronizer()
macronizer.settext(texttomacronize)
if scan == HEXAMETER:
macronizer.scan([Macronizer.dactylichexameter])
elif scan == ELEGIACS:
macronizer.scan([
Macronizer.dactylichexameter, Macronizer.dactylicpentameter
])
elif scan == HENDECA:
macronizer.scan([Macronizer.hendecasyllable])
elif scan == IAMBTRIDI:
macronizer.scan(
[Macronizer.iambictrimeter, Macronizer.iambicdimeter])
macronizedtext = macronizer.gettext(domacronize,
alsomaius,
performutov,
performitoj,
def create_html_page(scriptname, texttomacronize, domacronize, alsomaius, scan,
performitoj, performutov, doevaluate):
texttomacronize = unicodedata.normalize('NFC',
texttomacronize).replace('\r', '')
texttomacronize = texttomacronize[:TRUNCATETHRESHOLD]
if texttomacronize.startswith(DEBUGCOMMAND):
dodebug = True
texttomacronize = texttomacronize[len(DEBUGCOMMAND):]
else:
dodebug = False
errormessage = ""
if not texttomacronize:
macronizedtext = ""
else:
try:
macronizer = Macronizer()
macronizer.settext(texttomacronize)
if scan > 0:
macronizer.scan(SCANSIONS[scan][1])
macronizedtext = macronizer.gettext(domacronize,
alsomaius,
performutov,
performitoj,
markambigs=False)
except Exception as inst:
errormessage = inst.args[0]
macronizedtext = ""
html = []
html.append(
"""<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<style type="text/css">
h1 a {font-style: italic; text-decoration: none; color: black;}
body {padding: 2em;}
span.wrong {background-color: #ff6666;}
span.ambig {background-color: yellow;}
span.unknown {background-color: orange;}
span.fixed {background-color: lightgreen;}
div.feet {float: left; border: 1px hidden; padding: 0.5em;}
div.prewrap {display:inline-block; white-space: pre-wrap; border: 1px dashed black; padding: 0.5em;}
</style>
<title>A Latin Macronizer</title>
</head>
<body>
<h1><a href="%(scriptname)s">A Latin Macronizer</a></h1>
<p>Please enter a Latin text!</p>
%(truncatewarning)s
<form action="%(scriptname)s" method="post">
<p><textarea name="textcontent" onclick="enlarge(this)" cols="100" rows="%(numrows)s">%(errormessage)s</textarea><br>
<input type="checkbox" name="macronize" onchange="toggleDisabled(this.checked)" value="on" %(domacronize)s> Mark long vowels.<br>
<input class="macronizersetting" type="checkbox" name="alsomaius" value="on" %(alsomaius)s> Also mark <i>māius</i> etc.<br>
To improve the result, try to scan the text as <select name="scan">
%(scanoptions)s
</select>.<br>
<input class="macronizersetting" type="checkbox" name="doevaluate" value="off" %(doevaluate)s> Compare result with correctly macronized input text.<br>
<input type="checkbox" name="utov" value="on" %(performutov)s> Convert u to v.<br>
<input type="checkbox" name="itoj" value="on" %(performitoj)s> Convert i to j.<br>
<input type="submit" value="Submit"> (Please be patient!)<br>
</p></form>
""" % {
'scriptname':
scriptname,
'truncatewarning':
'' if TRUNCATETHRESHOLD < 0 else
'<p>Note: In order to avoid time out from the server, input longer than %s characters will be truncated. Sorry about that!</p>'
% TRUNCATETHRESHOLD,
'numrows':
20 if not texttomacronize else 3,
'errormessage':
errormessage,
'domacronize':
'checked' if domacronize else '',
'alsomaius':
'checked' if alsomaius else '',
'scanoptions':
''.join([
'<option value="%i"%s>%s</option>' %
(i, ' selected' if scan == i else '', description)
for i, [description, _] in enumerate(SCANSIONS)
]),
'doevaluate':
'checked' if doevaluate else '',
'performutov':
'checked' if performutov else '',
'performitoj':
'checked' if performitoj else '',
})
if macronizedtext:
html.append('<h2>Result</h2>')
html.append(
'<p>(Ambiguous forms are marked <span class="ambig">yellow</span>; unknown forms are <span class="unknown">orange</span>. You may click on a vowel to add or remove a macron.)</p>'
)
if scan > 0:
html.append('<div class="feet">%s</div>' %
'<br>'.join(macronizer.tokenization.scannedfeet))
html.append(
'<div class="prewrap" id="selectme" contenteditable="true">%s</div>'
% macronizer.tokenization.detokenize(True))
html.append(
'<p><input id="selecttext" type="button" value="Copy text"></p>')
if doevaluate:
html.append('<h2>Evaluation</h2>')
(accuracy, evaluatedtext) = evaluate(texttomacronize,
macronizedtext)
html.append('<div class="prewrap">%s</div>' % evaluatedtext)
html.append('<p>Accuracy: %f%%</p>' % (accuracy * 100))
if dodebug:
html.append('<h2>Debug info</h2>')
html.append('<pre>%s</pre>' % macronizer.tokenization.show())
html.append("""<h2>News</h2>
<p>August 2017: More meters added! The macronizer can now handle hendecasyllables as well as distichs of iambic trimeters and dimeters (<i>Beātus ille quī procul negōtiīs...</i>).</p>
<p>May 2017: I have now made the macronized text editable, which means that it will now be much easier to correct typos or misspellings while proofreading the text.</p>
<p>October 2016: The performance on texts written in all uppercase letters has been greatly improved.</p>
<p>July 2016: I am happy to announce that the Macronizer now is
able to take the meter into account when guessing the vowel
lengths in poetry. When tested on a couple of books of the
Aeneid (from the eminent <a href="http://dcc.dickinson.edu/">Dickinson College
Commentaries</a>), this has been demonstrated to cut the number of
erroneous vowel lengths in half! Currently, dactylic hexameters
and elegiac distichs are supported; other meters may be added.</p>
<p>Also, I have now added a PayPal donation button:
if you use the macronizer regularly and find it helpful and
time-saving, please consider making a donation, to support
maintenance and continuous development! Any amount is very much
appreciated!
<form action="https://www.paypal.com/cgi-bin/webscr" method="post" target="_top">
<p>
<input type="hidden" name="cmd" value="_s-xclick">
<input type="hidden" name="hosted_button_id" value="KKJ2V4ZVB3WGU">
<input type="image" src="https://www.paypalobjects.com/en_US/SE/i/btn/btn_donateCC_LG.gif" border="0" name="submit" alt="PayPal - The safer, easier way to pay online!">
<img alt="" border="0" src="https://www.paypalobjects.com/en_US/i/scr/pixel.gif" width="1" height="1">
</p>
</form>
<h2>Information</h2>
<p>This automatic macronizer lets you quickly mark all the long vowels
in a Latin text. The expected accuracy on an average classical text is
estimated to be about 98% to 99%. Please review the resulting macrons
with a critical eye!</p>
<p>The macronization is performed using a part-of-speech tagger
(<a href="http://www.cis.uni-muenchen.de/~schmid/tools/RFTagger/">RFTagger</a>)
trained on the <a href="http://www.dh.uni-leipzig.de/wo/projects/ancient-greek-and-latin-dependency-treebank-2-0/">Latin
Dependency Treebank</a>, and with macrons provided by a customized
version of the Morpheus morphological analyzer. An earlier version of
this tool was the subject of my bachelor’s thesis in Language
Technology, <i><a href="http://stp.lingfil.uu.se/exarb/arch/winge2015.pdf">Automatic
annotation of Latin vowel length</a></i>.</p>
<p>If you want to run the macronizer locally, or develop it further,
you may find the <a href="https://github.com/Alatius/latin-macronizer">source
code on GitHub</a>.</p>
<p>Copyright 2015-2021 Johan Winge. Please send comments to
<a href="mailto:[email protected]">[email protected]</a>.</p>
<script type="text/javascript">
function enlarge(textbox) {
textbox.rows = 20;
}
function toggleDisabled(_checked) {
var elements = document.getElementsByClassName('macronizersetting')
for (var i = 0; i < elements.length; i++) {
elements[i].disabled = _checked ? false : true;
}
}
function clickHandler(event) {
var span = event.target;
if (span.className == 'ambig' || span.className == 'unknown' || span.className == 'auto') {
span.className = 'fixed';
return;
}
if (span.parentNode.className == 'ambig' || span.parentNode.className == 'unknown' || span.parentNode.className == 'auto' || span.parentNode.className == 'fixed') {
span.parentNode.className = 'fixed';
if (span.innerHTML == 'ā') { span.innerHTML = 'a'; }
else if (span.innerHTML == 'a') { span.innerHTML = 'ā'; }
else if (span.innerHTML == 'ē') { span.innerHTML = 'e'; }
else if (span.innerHTML == 'e') { span.innerHTML = 'ē'; }
else if (span.innerHTML == 'ī') { span.innerHTML = 'i'; }
else if (span.innerHTML == 'i') { span.innerHTML = 'ī'; }
else if (span.innerHTML == 'ō') { span.innerHTML = 'o'; }
else if (span.innerHTML == 'o') { span.innerHTML = 'ō'; }
else if (span.innerHTML == 'ū') { span.innerHTML = 'u'; }
else if (span.innerHTML == 'u') { span.innerHTML = 'ū'; }
else if (span.innerHTML == 'ȳ') { span.innerHTML = 'y'; }
else if (span.innerHTML == 'y') { span.innerHTML = 'ȳ'; }
else if (span.innerHTML == 'Ā') { span.innerHTML = 'A'; }
else if (span.innerHTML == 'A') { span.innerHTML = 'Ā'; }
else if (span.innerHTML == 'Ē') { span.innerHTML = 'E'; }
else if (span.innerHTML == 'E') { span.innerHTML = 'Ē'; }
else if (span.innerHTML == 'Ī') { span.innerHTML = 'I'; }
else if (span.innerHTML == 'I') { span.innerHTML = 'Ī'; }
else if (span.innerHTML == 'Ō') { span.innerHTML = 'O'; }
else if (span.innerHTML == 'O') { span.innerHTML = 'Ō'; }
else if (span.innerHTML == 'Ū') { span.innerHTML = 'U'; }
else if (span.innerHTML == 'U') { span.innerHTML = 'Ū'; }
else if (span.innerHTML == 'Ȳ') { span.innerHTML = 'Y'; }
else if (span.innerHTML == 'Y') { span.innerHTML = 'Ȳ'; }
}
}
function attachHandler(container) {
if (container.addEventListener) {
container.addEventListener('click', clickHandler, false);
} else if (container.attachEvent) {
container.attachEvent('onclick', function(e) {
return clickHandler.call(container, e || window.event);
});
}
}
var ambigs = document.getElementsByClassName("ambig"),
unknowns = document.getElementsByClassName("unknown");
autos = document.getElementsByClassName("auto");
for (var i = 0; i < ambigs.length; i++) {
attachHandler(ambigs[i]);
}
for (var i = 0; i < unknowns.length; i++) {
attachHandler(unknowns[i]);
}
for (var i = 0; i < autos.length; i++) {
attachHandler(autos[i]);
}
document.getElementById("selecttext").onclick = function () {
var text = document.getElementById("selectme"), range, selection;
if (document.body.createTextRange) {
range = document.body.createTextRange();
range.moveToElementText(text);
range.select();
} else if (window.getSelection) {
selection = window.getSelection();
range = document.createRange();
range.selectNodeContents(text);
selection.removeAllRanges();
selection.addRange(range);
}
var successful = document.execCommand('copy');
};
</script>
</body>
</html>""")
return '\n'.join(html)