*
* This file is a part of slate_permutate.
*
* slate_permutate is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* slate_permutate is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Affero General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with slate_permutate. If not, see .
*/
define('CCBCMD_CRAWL_URI', 'http://ccbcmd.edu/schedule/sched.html');
/**
* \brief
* Obtain list of crawlable semesters offered by CCBCMD.
*
* \parram $school
* The CCBCMD school handle.
* \param $semesters
* Array to populate with available semesters.
* \return
* 0 on success.
*/
function ccbcmd_crawl_semester_list(array $school, array &$semesters, &$school_crawl_log)
{
$cookies = array();
/*
* It seems that http://ccbcmd.edu/schedule/sched.html is what we're
* meant to start from. That's just a redirect to some other page
* from which we get a listing of available semesters and choose
* one.
*/
$uri = CCBCMD_CRAWL_URI;
$semesters_dom = new DOMDocument();
$semesters_dom->loadHTML(school_crawl_geturi($uri, $cookies, $school_crawl_log, NULL, TRUE, 'ccbcmd_crawl_curlhook'));
$semesters_select_node = $semesters_dom->getElementById('term_input_id');
if ($semesters_select_node === NULL)
{
school_crawl_logf($school_crawl_log, 0, "Could not get list of available semesters to choose from.");
return 1;
}
foreach ($semesters_select_node->childNodes as $semesters_option_node)
{
$semester_text = $semesters_option_node->textContent;
$semester_value = $semesters_option_node->getAttribute('value');
if (empty($semester_value))
/* skip the empty ``None'' semester */
continue;
if (stripos($semester_text, 'continuing') !== FALSE)
/* skip the year-long semesters dedicated to continuing education */
continue;
list($semester_season, $semester_year) = explode(' ', $semester_text);
/* the college has two separate summer sessions, so distinguish between them */
if (preg_match(';session ([0-9]+);i', $semester_text, $matches))
$semester_season .= '_' . $matches[1];
$semesters[] = new Semester($semester_year, strtolower($semester_season));
}
return 0;
}
/**
* \brief
* Crawl a CCBCMD semester.
*
* \param $school
* The CCBCMD school handle.
* \param $semester
* The semester to fill with courses.
*/
function ccbcmd_crawl_semester($school, $semester, &$school_crawl_log)
{
$cookies = array();
$uri = CCBCMD_CRAWL_URI;
$semesters_dom = new DOMDocument();
$semesters_dom->loadHTML(school_crawl_geturi($uri, $cookies, $school_crawl_log, NULL, TRUE, 'ccbcmd_crawl_curlhook'));
$semesters_select_node = $semesters_dom->getElementById('term_input_id');
if (empty($semesters_select_node))
{
school_crawl_logf($school_crawl_log, 0, "Could not locate the list of semesters from which to choose.");
return 1;
}
$semesters_form = school_crawl_element_ancestor($semesters_select_node, 'form');
if ($semesters_form === NULL)
{
school_crawl_logf($school_crawl_log, 0, "Unable to find
associated with semester.");
return 1;
}
$semesters_post = school_crawl_form($semesters_form);
$semester_found = FALSE;
foreach ($semesters_select_node->childNodes as $semesters_option_node)
{
$semester_text = $semesters_option_node->textContent;
$semester_value = $semesters_option_node->getAttribute('value');
if (empty($semester_value))
continue;
list($semester_season, $semester_year) = explode(' ', $semester_text);
if (preg_match(';session ([0-9]+);i', $semester_text, $matches))
$semester_season .= '_' . $matches[1];
$semester_season = strtolower($semester_season);
if ($semester_year == $semester->year_get()
&& $semester_season == $semester->season_get())
{
$semester_found = TRUE;
break;
}
}
if (!$semester_found)
{
school_crawl_logf($school_crawl_log, 1, "Unable to find the entry for semester %s.", $semester);
return 1;
}
$semesters_post[$semesters_select_node->getAttribute('name')] = $semester_value;
$subjects_dom = new DOMDocument();
$uri = school_crawl_url($uri, $semesters_form->getAttribute('action'));
$subjects_dom->loadHTML(school_crawl_geturi($uri, $cookies, $school_crawl_log, $semesters_post, TRUE, 'ccbcmd_crawl_curlhook'));
$subjects_form_nodelist = $subjects_dom->getElementsByTagName('form');
if (!$subjects_form_nodelist->length)
{
school_crawl_logf($school_crawl_log, 0, "Unable to find to submit for the subjects-choosing page.");
return 1;
}
$subjects_form_node = $subjects_form_nodelist->item(0);
$subjects_post = school_crawl_form($subjects_form_node);
$subjects_select_node = $subjects_dom->getElementById('subj_id');
foreach ($subjects_select_node->childNodes as $subjects_option_node)
if (!strcasecmp('all', trim($subjects_option_node->textContent)))
$subjects_post[$subjects_select_node->getAttribute('name')][] = $subjects_option_node->getAttribute('value');
$courses_dom = new DOMDocument();
$uri = school_crawl_url($uri, $subjects_form_node->getAttribute('action'));
$courses_dom->loadHTML(school_crawl_geturi($uri, $cookies, $school_crawl_log, $subjects_post, TRUE, 'ccbcmd_crawl_curlhook'));
$courses_xpath = new DOMXPath($courses_dom);
/* The second row of the table has all of the headers in it */
$tr_header_nodelist = $courses_xpath->query('//table[@class="datadisplaytable" and position()=1]//tr[position()=2]');
if (!$tr_header_nodelist->length)
{
school_crawl_logf($school_crawl_log, 0, "Unable to find the row of the course/section data table which gives us the mappings of column names onto columns.");
return 1;
}
$tr_header_node = $tr_header_nodelist->item(0);
$section_offsets = array(
'registration_number' => school_crawl_table_resolve_column($tr_header_node, 'CRN'),
'section_id' => school_crawl_table_resolve_column($tr_header_node, 'subj/crse/sec'),
/* there's a boolean column which says whether or not the course has any prerequisites/corequisites.... */
'credits' => school_crawl_table_resolve_column($tr_header_node, 'credhrs'),
/* there's a column for the number of contact hours, vs. credit hours */
'dates' => school_crawl_table_resolve_column($tr_header_node, 'sessiondates'),
);
foreach (array('title', 'days', 'times', 'instructor', 'location') as $column_key)
$section_offsets[$column_key] = school_crawl_table_resolve_column($tr_header_node, $column_key);
/* there's also a column for ``session dates'' */
/* error check and calculate the number of children that a node must have to be */
$max_offset = 0;
foreach ($section_offsets as $name => $value)
{
if ($value === FALSE)
{
school_crawl_logf($school_crawl_log, 0, "Unable to find column offset for `%s'.",
$name);
return 1;
}
else
school_crawl_logf($school_crawl_log, 9, "%s -> %s", $name, $value);
$max_offset = max($max_offset, $value);
}
foreach ($courses_xpath->query('//table[@class="datadisplaytable" and position()=1]//tr') as $tr_node)
{
$children = school_crawl_table_rownodes($tr_node);
if ($children->length < $max_offset)
/*
* Skip this row because it doesn't have all of the columns we
* want and thus it can't be a row containing information
* about a section.
*/
continue;
if (!strcmp($children->item($section_offsets['section_id'])->tagName, 'th'))
/*
* We've hit one of the
s filled with | s. Skip this one.
*/
continue;
/*
* There are some rows with the time set to TBA and with empty
* section_id columns. Respond to this by skipping empty
* section_id columns since there's no useful data in these
* rows. We use strlen() < 3 because trim() doesn't take care of
* :-/
*
* There are other times that the section_id row is empty and
* the time column is set to something. In this case, the
* subsequent rows are describing additional SectionMeetings
* which should be added to the existing Section.
*/
$section_id = trim($children->item($section_offsets['section_id'])->textContent);
if (strlen($section_id) > 2)
{
/**
* \todo
* If a section's section ID ends in `W', like `EFW', that
* means it's a semi-online course. We should probably
* distinguish these from normal sections, probably
* disabling them from showing up by default.
*/
$section_id_parts = Section::parse($section_id);
$registration_number = $children->item($section_offsets['registration_number'])->textContent;
$credit_hours = (float)$children->item($section_offsets['credits'])->textContent;
$section = new Section($section_id_parts['section'], array(), $registration_number, $credit_hours);
$semester->section_add($section_id_parts['department'], $section_id_parts['course'], $section,
trim($children->item($section_offsets['title'])->textContent));
}
if (empty($section))
{
school_crawl_logf($school_crawl_log, 4, "Expected a section row beofre having a row with only partial data. Ignoring row.");
continue;
}
$instructor = $children->item($section_offsets['instructor'])->textContent;
{
$time_range_text = $children->item($section_offsets['times'])->textContent;
if (strpos($time_range_text, 'TBA') !== FALSE)
/*
* There is no way to get meeting info and create
* SectionMeetings.
*/
continue;
if (($dash_pos = strpos($time_range_text, '-')) === FALSE)
{
school_crawl_logf($school_crawl_log, 0, "Unable to understand course's time range format, cannot find dash: ``%s''.",
$time_range_text);
return 1;
}
$time_start_text = substr($time_range_text, 0, $dash_pos);
$time_start = strptime($time_start_text, '%I:%M %p');
$time_end_text = substr($time_range_text, $dash_pos + 1);
/*
* Make sure that _only_ one date range is specified to ensure
* data integrity. I.e., make sure that the college doesn't
* suddenly support multiple meeting times in one field
* without our anticipating that and then cause us to have
* invalid data. ;-). The college does support multiple
* section meetings, it does this by having multiple rows per
* section. The extra rows _only_ have the days, time, prof,
* and dates columns. --binki
*/
if (strpos($time_end_text, '-') !== FALSE)
{
school_crawl_logf($school_crawl_log, 4, "Entry seems to have invalid date column data: ``%s'' time_end_text: ``%s''.",
$time_range_text, $time_end_text);
continue;
}
$time_end = strptime($time_end_text, '%I:%M %p');
if ($time_end === FALSE || $time_start === FALSE)
{
school_crawl_logf($school_crawl_log, 4, "Error parsing start or end time: start: ``%s'' end: ``%s''.",
$time_start_text, $time_end_text);
continue;
}
$days = school_crawl_days_str_format($school_crawl_log, $children->item($section_offsets['days'])->textContent);
$section->meeting_add(new SectionMeeting($days, school_crawl_time_format($time_start), school_crawl_time_format($time_end),
$children->item($section_offsets['location'])->textContent,
'lecture',
$instructor));
/* check if a semester's date range should be increased */
$section_dates = $children->item($section_offsets['dates'])->textContent;
if (preg_match(';^([0-9]+)/([0-9]+)-([0-9]+)/([0-9]+)$;', $section_dates, $section_dates_matches))
{
$semester->time_start_set_test(gmmktime(0, 0, 0, $section_dates_matches[1], $section_dates_matches[2], $semester->year_get()));
$semester->time_end_set_test(gmmktime(0, 0, 0, $section_dates_matches[3], $section_dates_matches[4], $semester->year_get()));
}
}
}
return 0;
}
function ccbcmd_crawl_curlhook(&$curl)
{
/*
* OK, so this must be set to SSLv2 or SSLv3 because of how the
* server's SSL junk is messed up. When curl is built against
* gnutls, though, we can't use SSL2 since it doesn't support that
* old of a protocol. So, we use 3 which works. Apparently, the
* server can't handle gnutls's attempt to use TLS. Even openssl's
* s_client command fails without manually specifying --ssl2 or
* --ssl3. So, this must be a _really_ weird server setup...
*/
curl_setopt($curl, CURLOPT_SSLVERSION, 3);
}