*
* 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 .
*/
/**
* \file
*
* A crawler for the WebAdvisor webapp.
*/
$incdir = dirname(__FILE__) . DIRECTORY_SEPARATOR;
require_once $incdir . 'class.semester.inc';
require_once $incdir . 'class.course.inc';
require_once $incdir . 'class.section.php';
require_once $incdir . 'class.section_meeting.inc';
define('_SCHOOL_CRAWL_WEBADVISOR_START_FORM', '?type=P&pid=ST-WESTS12A&LASTTOKEN=NULL');
function _school_crawl_webadvisor_common_prep(array &$school, array &$options)
{
$school += array('webadvisor_url' => $school['url'] . 'WebAdvisor');
$options += array(
'season_mapper' => 'school_crawl_webadvisor_season_mapper',
'curlsetup_hook' => NULL,
);
}
/*
* \brief
* Crawl the list of semesters available from a
* WebAdvisor-compatible school.
*
* \param $school
* The school’s info array/handle.
* \param $semesters
* The array to populate with various semesters available at this
* college.
* \param $school_crawl_log
* The school_crawl_log handle.
* \param $season_mapper
* A function($term, $term_value) which maps term names onto
* semester/season names. See
* school_crawl_webadvisor_season_mapper() for the default
* implementation.
*/
function school_crawl_webadvisor_semester_list(array $school, array &$semesters, &$school_crawl_log, array $options = array())
{
_school_crawl_webadvisor_common_prep($school, $options);
$cookies = array();
$uri = $school['webadvisor_url'] . _SCHOOL_CRAWL_WEBADVISOR_START_FORM;
$semesters_html = school_crawl_webadvisor_geturi($uri, $cookies, $school_crawl_log, $options);
$semesters_dom = new DOMDocument();
$semesters_dom->loadHTML($semesters_html);
/*
* Discover the available semesters
*/
$semesters_var1 = $semesters_dom->getElementById('VAR1');
if (empty($semesters_var1))
{
school_crawl_logf($school_crawl_log, 0, "Error: Unable to load list of semesters.");
return 1;
}
$semesters_select_nodes = $semesters_var1->childNodes;
foreach ($semesters_select_nodes as $semester_node)
{
if ($semester_node->tagName != 'option'
|| !$semester_node->hasAttribute('value')
|| !strlen($semester_node->getAttribute('value')))
continue;
$term = $semester_node->textContent;
$term_value = $semester_node->getAttribute('value');
$semester = $options['season_mapper']($term, $term_value, $school_crawl_log);
/*
* We need a way to map a semester back to a list of
* term_values. We can tack an extra member variable onto any
* object in PHP, so we use that method.
*/
if (!empty($semester))
if (empty($semesters[$semester->id()]))
{
$semester->_school_crawl_webadvisor_term_values = array($term_value);
$semesters[$semester->id()] = $semester;
}
else
/*
* A semester associated with this year/season already
* exists. Append an additional term value to be associated
* with this Semester so that they can be aggregated when
* crawled later.
*/
$semesters[$semester->id()]->_school_crawl_webadvisor_term_values[] = $term_value;
}
return 0;
}
function school_crawl_webadvisor_geturi(&$uri, array &$cookies, &$school_crawl_log, array $options)
{
/**
* We have to handle the case where the user is first browing to
* WebAdvisor. For example, with the ST-WESTS12A sequence:
*
* Start the ST-WESTS12A sequence.
*
* 1. WebAdvisor?type=P&pid=ST-WESTS12A&LASTTOKEN=NULL
*
*
* Calls javascript:getWindowHTML(). This merely adds
* TOKENIDX=NULL to the query string, so we can skip this step
* and just have TOKENIDX=NULL.
*
* 2. WebAdvisor?type=P&pid=ST-WESTS12A&LASTTOKEN=NULL&TOKENIDX=NULL
*
*
* 3. WebAdvisor?type=P&pid=ST-WESTS12A&TOKENIDX=7699844013 In #2,
* the second argument to setWindowHTML() is random. Thus, we
* have to capture this value and set it as GET parameter named
* “TOKENIDX”.
*/
if (strpos($uri, 'TOKENIDX') === FALSE)
{
if (strpos($uri, '?') === FALSE)
$uri .= '?';
else
$uri .= '&';
/* Starting value. */
$uri .= 'TOKENIDX=NULL';
}
$html = school_crawl_webadvisor_noscript_filter(school_crawl_geturi($uri, $cookies, $school_crawl_log, NULL, FALSE, $options['curlsetup_hook']));
if (!preg_match('/setWindowHTML\\(\'\', \'([0-9]+)\'\\);/', $html, $matches))
/*
* The user already had a valid TOKENIDX, so we’re good to go.
*/
return $html;
$token = $matches[1];
school_crawl_logf($school_crawl_log, 7, "Using WebAdvisor TOKENIDX=%s.", $token);
/*
* setWindowHTML() will first remove the query string parameters
* 'CLONE' and 'FORCEIDX'. Then it appends TOKENIDX= to the
* query parameters.
*
* Example, where TOKENIDX does not start out as NULL but where a
* CLONE=Y command is being sent:
*
* Input: HTTPS://kvdata.calvin.edu/walive/WebAdvisor?TYPE=P&PID=ST-WESTS13C&CLONE=Y&CLONE_PROCESS=Y&SPAUKQ=708501792841963&TOKENIDX=1507971558
*
* Result: HTTPS://kvdata.calvin.edu/walive/WebAdvisor?TYPE=P&PID=ST-WESTS13C&CLONE_PROCESS=Y&SPAUKQ=708501792841963&TOKENIDX=2281086932
*/
$uri = preg_replace('/([?&])TOKENIDX=[^&]+/', '$1TOKENIDX=' . $token,
preg_replace('/([?&])(CLONE|FORCEIDX)=[^&]+&?/', '$1', $uri));
return school_crawl_webadvisor_noscript_filter(school_crawl_geturi($uri, $cookies, $school_crawl_log, NULL, FALSE, $options['curlsetup_hook']));
}
/**
* \brief
* Searches for and removes a element.
*
* The WebAdvisor likes to put in a docs and
* place things other than , , and in this
* . This is invalid HTML
* (http://www.w3.org/TR/html5/the-noscript-element.html#the-noscript-element)
* and not handled by libxml2’s DOM.
*
* \param $html
* The input HTML to filter.
* \return
* The fixed HTML.
*/
function school_crawl_webadvisor_noscript_filter($html)
{
return preg_replace(';\<(noscript)\>.*?\;s', '', $html);
}
/**
* \brief
* Map a term name onto a season and generate the appropriate
* Semester object.
*
* \param $term
* The human-friendly term.
* \param $term_value
* The form value of this term.
* \return
* NULL if unable to generate the Semester, otherwise an empty
* Semester object with its year and season set.
*/
function school_crawl_webadvisor_season_mapper($term, $term_value, &$school_crawl_log)
{
if (!preg_match('/(^|[^\d])(\d{4})($|[^\d])/', $term, $matches))
{
school_crawl_logf($school_crawl_log, 2, "Unable to interpret “%s” with form value of “%s” as specifying a particular year.",
$term, $term_value);
return NULL;
}
$year = $matches[2];
$term_minusyear = trim(str_replace($year, '', $term));
list($season) = explode(' ', strtolower($term_minusyear));
if (empty($season))
{
school_crawl_logf($school_crawl_log, 2, "Unable to interpret “%s” with form value of “%s” as specifying a particular season.",
$term, $term_value);
return NULL;
}
/*
* Try to coerce into a valid season name. For example, will coerce
* 'sp' into 'spring'.
*/
$season_strlen = strlen($season);
foreach (Semester::seasons_get_all() as $valid_season)
if (!strncmp($valid_season, $season, min(strlen($valid_season), $season_strlen)))
$season = $valid_season;
school_crawl_logf($school_crawl_log, 9, "Interpreting “%s” term as %s-%s.", $term, $year, $season);
return new Semester($year, $season);
}
/**
* \brief
* Map a particular Semester onto a list of term_values for using as
* “VAR1” in the initial schedule form.
*
* \param $school
* The school handle.
* \param $semester
* The Semester to map onto a list of term_values.
* \param $school_crawl_log
* The school_crawl_log handle.
* \param $season_mapper
* The season_mapper used with
* school_crawl_webadvisor_semester_list().
* \return
* An array, possibly empty, of term_values associated with this
* Semester or NULL on error.
*/
function _school_crawl_webadvisor_semester_toterms(array $school, Semester $semester, &$school_crawl_log, array $options)
{
if (!empty($semester->_school_crawl_webadvisor_term_values))
return $semester->_school_crawl_webadvisor_term_values;
$semesters = array();
$ret = school_crawl_webadvisor_semester_list($school, $semesters, $school_crawl_log, $options);
if ($ret)
{
school_crawl_logf($school_crawl_log, 4, "Unable to map Semester “%s” onto a term_value because crawling the semester list failed.", $semester);
return NULL;
}
if (empty($semesters[$semester->id()]))
{
school_crawl_logf($school_crawl_log, 4, "Unable to map Semester “%s” onto a semester because no matching semester was found in the semester list.", $semester);
return NULL;
}
return $semesters[$semester->id()]->_school_crawl_webadvisor_term_values;
}
/**
* \brief
* Crawl the courses for a semester from a WebAdvisor instance.
*
* There may be multiple terms associated with a particular
* semester. For example,
* https://solomon.cornerstone.edu/COLLIVE/WebAdvisor?CONSTITUENCY=WBST&type=P&pid=ST-WESTS12A
* lists overlapping terms such as “Spring 2013 Undergraduate”,
* “Spring 2013 Seminary”, and “SP 2013 GRD (MAML/MABS)”.
*
* \param $school
* The school handle.
* \param $semester
* The Semester to crawl.
* \param $school_crawl_log
* The school_crawl_log handle.
* \param $season_mapper
* The season_mapper used with
* school_crawl_webadvisor_semester_list() if any.
*/
function school_crawl_webadvisor_semester(array $school, Semester $semester, &$school_crawl_log, array $options = array())
{
_school_crawl_webadvisor_common_prep($school, $options);
$cookies = array();
$uri = $school['webadvisor_url'] . _SCHOOL_CRAWL_WEBADVISOR_START_FORM;
$html = school_crawl_webadvisor_geturi($uri, $cookies, $school_crawl_log, $options);
$form_uri = $uri;
$seed_dom = new DOMDocument();
$seed_dom->loadHTML($html);
$seed_form = _school_crawl_webadvisor_form($seed_dom, $school_crawl_log);
if (empty($seed_form))
return 1;
$return_url = reset($seed_form['RETURN.URL']);
/*
* First, read all of the friendly subject/department names. They're
* not in the output, but they're in the “Subjects” dropdown of
* the input form. The
* is associated with subjects/departments.
*/
$department_var1_list = array();
foreach (school_crawl_form_select_array($seed_dom->getElementById('LIST_VAR1_1')) as $department_id => $department_name)
{
$semester->department_name_set($department_id, trim(reset($department_name)));
$department_var1_list[] = $department_id;
}
foreach (_school_crawl_webadvisor_semester_toterms($school, $semester, $school_crawl_log, $options) as $semester_str)
while (count($department_var1_list))
{
/* Start back on the form page... */
$uri = $form_uri;
/*
* LIST.VAR_: is the column, is the row. There
* are apparently a max of 5 rows (see the LIST.VAR_MAX
* below).
*
* Columns:
* LIST.VAR1: department
* LIST.VAR2: course_level
* LIST.VAR3: IIRC, a course identifier, such as 156 from MATH-156
* LIST.VAR4: I forget
*
*/
school_crawl_logf($school_crawl_log, 6, 'Using %s for a semester form value.',
$semester_str);
$form = array('VAR1' => $semester_str,
'LIST.VAR1_1' => '',
'LIST.VAR2_1' => '',
/*
* Other form items we're not querying but which need
* to be sent blankly.
*/
'RETURN.URL' => $return_url,
'SUBMIT_OPTIONS' => '',
/*
* The submit button... its value="" key is
* apparently sent with the form... makes a
* little bit of sense I guess ;-).
*/
/*'SUBMIT2' => 'SUBMIT',*/
'DATE.VAR1' => '',
'DATE.VAR2' => '',
'LIST.VAR1_CONTROLLER' => 'LIST.VAR1',
'LIST.VAR1_MEMBERS' => 'LIST.VAR1*LIST.VAR2*LIST.VAR3*LIST.VAR4',
);
foreach (array('1', '2', '3', '4') as $list_col)
{
$colname = 'LIST.VAR' . $list_col;
if (!isset($form[$colname . '_MAX']))
$form[$colname . '_MAX'] = '5';
foreach (array('1', '2', '3', '4', '5') as $list_row)
{
$rowname = $colname . '_' . $list_row;
if (!isset($form[$rowname]))
$form[$rowname] = '';
}
}
/*
* Fill in some semesters.
*/
foreach (array('1', '2', '3', '4', '5') as $var1_row)
if (count($department_var1_list))
{
$form['LIST.VAR1_' . $var1_row] = array_shift($department_var1_list);
}
/*
* VAR7 and VAR 8 is a constraint of times during which
* courses meet
*/
$form['VAR7'] = '';
$form['VAR8'] = '';
/* “course title keywords” */
$form['VAR3'] = '';
/* ? */
$form['VAR6'] = '';
$form['VAR21'] = '';
/* instructor's last name */
$form['VAR9'] = '';
/*
* VAR10 through VAR16 are Monday through Sunday checkboxes
* for days of the week that classes meet.
*
* But we specify no days of the week to avoid this being a
* constraint ;-).
*/
/*
for ($day = 10; $day <= 16; $day ++)
$form['VAR' . $day] = '';
*/
$skipped_sections = array('incomplete meeting info' => 0, 'invalid meeting info format' => 0);
/*
* pages is populated by preg_match() below after the first looping.
*/
$pages = array(1 => 0, 2 => 1);
while ($pages[1] < $pages[2])
{
$html = school_crawl_webadvisor_noscript_filter(school_crawl_geturi($uri, $cookies, $school_crawl_log, $form, FALSE, $options['curlsetup_hook']));
$results_dom = new DOMDocument();
$results_dom->loadHTML($html);
$results_form = _school_crawl_webadvisor_form($results_dom, $school_crawl_log);
if (empty($results_form))
return 1;
$list_done = FALSE;
for ($list_row = 1; !$list_done; $list_row ++)
{
/* either 'Open' (or 'Closed'?) */
$openness = empty($results_form['LIST.VAR1_' . $list_row]) ? NULL : reset($results_form['LIST.VAR1_' . $list_row]);
$sec_short_title = _school_crawl_webadvisor_dom_id_content($results_dom, 'SEC_SHORT_TITLE_' . $list_row);
$sec_meetings_info = _school_crawl_webadvisor_dom_id_content($results_dom, 'SEC_MEETING_INFO_' . $list_row);
/* check if we're done with this particular page */
if (!strlen($openness) && !strlen($sec_short_title) && !strlen($sec_meetings_info))
{
$list_done = TRUE;
break;
}
/*
* The same info below should be retrievable with
* _school_crawl_webadvisor_dom_id_content($results_dom, 'SEC_FACULTY_INFO_' . $list_row);
*/
$faculty_name = reset($results_form['SEC.FACULTY.INFO_' . $list_row]);
$credits = reset($results_form['SEC.MIN.CRED_' . $list_row]); /* or id="SEC_FACULTY_INFO_$list_row" */
$comment = _school_crawl_webadvisor_dom_id_content($results_dom, 'SEC_COMMENTS_' . $list_row); /* or name="SEC.COMMENTS_$list_row" */
$short_title_onclick = $results_dom->getElementById('SEC_SHORT_TITLE_' . $list_row)->getAttribute('onclick');
/* parse */
$section_id = Section::parse($sec_short_title);
$synonym = NULL;
$title = NULL;
if (preg_match(';\(([0-9]+)\)(.*);', $sec_short_title, $matches))
{
$synonym = $matches[1];
$title = trim($matches[2]);
}
else
{
/*
* Places without synonyms just don’t have the
* parantheses around the synonym. Just get a title…
*/
list(, $title) = explode(' ', $sec_short_title, 2);
$title = trim($title);
}
school_crawl_logf($school_crawl_log, 10, "");
school_crawl_logf($school_crawl_log, 10, implode('-', $section_id) . ': ' . $sec_short_title);
school_crawl_logf($school_crawl_log, 10, $openness);
school_crawl_logf($school_crawl_log, 10, $sec_meetings_info);
school_crawl_logf($school_crawl_log, 10, $faculty_name);
school_crawl_logf($school_crawl_log, 10, $credits);
school_crawl_logf($school_crawl_log, 10, $comment);
school_crawl_logf($school_crawl_log, 10, "synonym: %s", $synonym);
school_crawl_logf($school_crawl_log, 10, "title: %s", $title);
/*
* The input format for this is, thankfully, pretty rigid
* :-D. Example input format:
*
* '01/31/2011-05/11/2011 Lecture Monday, Wednesday 01:00PM - 03:50PM, Spoelhof Center, Room 101'
*
* OR
*
* '01/31/2011-05/18/2011 Practicum Days to be Announced, Times to be AnnouncedTo Be Arranged, Room TBA'
*
* OR
*
* '01/31/2011-05/12/2011 Music Ensemble Monday, Wednesday, Thursday, Friday 03:30PM - 04:20PM, Covenant Fine Arts Center, Room 135'
*
* OR, per
* https://protofusion.org/bugzilla/show_bug.cgi?id=109 , we
* must parse the following on the main listing page and
* then parse more on the “course details” page:
*
* '09/06/2011-12/16/2011 Lecture Tuesday, Wednesday, Friday 12:30PM - 01:20PM, Science Building, Room 276 (more)...'
*
* The more on the “course details” page:
*
* '09/06/2011-12/16/2011 Lecture Tuesday, Wednesday, Friday 12:30PM - 01:20PM, Science Building, Room 276 09/06/2011-12/16/2011 Lecture Thursday 10:30AM - 12:20PM, Science Building, Room 276'
*
* Looks like in this last case parsing from right-to-left
* will be best.
*
* In the second case.... we'll just ignore the section. In
* the third case, we have to be careful about parsing out
* Monday.
*
* At this point, we don't parse most tokens. We group them
* off. We get the first date, the second date, the type
* ('Lecture', 'Practicum', or some other unknown value),
* the list of days of week the section meets, the start
* time, the end time, and then the meeting location.
*/
if (strpos($sec_meetings_info, 'Times to be Announced') !== FALSE
|| strpos($sec_meetings_info, 'Days to be Announced') !== FALSE)
{
school_crawl_logf($school_crawl_log, 8, 'Skipping class because of incomplete meeting time information: '
. implode('-', $section_id) . ' has meeting info of `'
. $sec_meetings_info . '\'');
$skipped_sections['incomplete meeting info'] ++;
/* Still add to have less confusing autocomplete */
school_crawl_webadvisor_course_add($semester, $section_id['department'], $section_id['course'], $title);
continue;
}
/*
* Check whether or not we have to pursue details on the
* “course detail page”. If we do, we might as well just
* parse the line of information available there instead of
* the same from the main listing page.
*/
if (preg_match('; \\(more\\)...$;', $sec_meetings_info)
&& preg_match(';^javascript:window\\.open\\(\'(.*?[^\\\\])\',;', $short_title_onclick, $short_title_onclick_matches))
{
$more_details_url = $short_title_onclick_matches[1];
$more_details_uri = strstr($uri, '?', TRUE) . $more_details_url;
school_crawl_logf($school_crawl_log, 8, 'Fetching extra course information page for %s-%s-%s from %s.',
$section_id['department'], $section_id['course'], $section_id['section'],
$more_details_uri);
$more_details_html = school_crawl_webadvisor_geturi($more_details_uri, $cookies, $school_crawl_log, $options);
$more_details_dom = new DOMDocument();
$more_details_dom->loadHTML($more_details_html);
/* Hopefully 'LIST_VAR12_1' is pretty constant... */
foreach ($more_details_dom->getElementById('LIST_VAR12_1')->childNodes as $more_details_child)
{
if ($more_details_child->nodeType != XML_TEXT_NODE)
continue;
$sec_meetings_info = $more_details_child->wholeText;
break;
}
school_crawl_logf($school_crawl_log, 9, "Result of fetching additional meeting information on next line(s):\n%s",
$sec_meetings_info);
}
/*
* If we have a course with multiple section_meetings, then
* $sec_meetings_info is split into each meeting by a
* "\n"
*/
foreach (explode("\n", $sec_meetings_info) as $sec_meeting_info)
{
if (!preg_match(';^([0-9]{2}/[0-9]{2}/[0-9]{4})-([0-9]{2}/[0-9]{2}/[0-9]{4}) (([^ ,]+ )+)([^0-9]+) ([^ ]+) - ([^ ]+), (.*)$;', $sec_meeting_info, $meeting_info_matches))
{
school_crawl_logf($school_crawl_log, 8, 'Unable to parse section meeting info string into start/end/days information for '
. implode('-', $section_id) . ': “' . $sec_meeting_info . '”');
$skipped_sections['invalid meeting info format'] ++;
/*
* Still add at least the course to the semester so that
* it shows up in autocmoplete.
*/
school_crawl_webadvisor_course_add($semester, $section_id['department'], $section_id['course'], $title);
continue;
}
$date_start = $meeting_info_matches[1];
$date_end = $meeting_info_matches[2];
/* e.g., 'Lecture', 'Practicum' */
$meeting_type = school_crawl_meeting_type($meeting_info_matches[3]);
$days = school_crawl_days_format($school_crawl_log, explode(', ', $meeting_info_matches[5]));
$time_start = school_crawl_time_format(strptime($meeting_info_matches[6], '%I:%M%p'));
$time_end = school_crawl_time_format(strptime($meeting_info_matches[7], '%I:%M%p'));
$meeting_place = $meeting_info_matches[8];
foreach (array('date_start', 'date_end', 'meeting_type', 'days', 'time_start', 'time_end', 'meeting_place', 'meeting_type') as $var)
school_crawl_logf($school_crawl_log, 10, "%s:%s", $var, ${$var});
$date_start_time = strptime($date_start, '%m/%d/%Y');
$date_end_time = strptime($date_end, '%m/%d/%Y');
if ($date_start_time !== FALSE)
$date_start_time = school_crawl_gmmktime($date_start_time, -5 * 60*60);
else
$date_start_time = NULL;
if ($date_end_time !== FALSE)
$date_end_time = school_crawl_gmmktime($date_end_time, -5 * 60*60) + 24*60*60;
else
$date_end_time = NULL;
$semester->section_meeting_add($section_id['department'], $section_id['course'], $title, $section_id['section'], $synonym,
new SectionMeeting($days, $time_start, $time_end, $meeting_place, $meeting_type, $faculty_name, $date_start_time, $date_end_time), 'default', $credits);
}
}
if (!preg_match(';Page ([0-9]+) of ([0-9]+)\\;', $html, $pages))
{
school_crawl_logf($school_crawl_log, 0, 'Unable to determine the number of pages in this resultset');
break;
}
school_crawl_logf($school_crawl_log, 8, "%s(): finished page %d of %d with %d courses.", __FUNCTION__, $pages[1], $pages[2], $list_row - 1);
$form = array(
'ACTION*Grp:WSS.COURSE.SECTIONS' => 'NEXT',
);
}
}
$has_stat = FALSE;
foreach ($skipped_sections as $reason => $num)
{
if (!$num)
continue;
if (!$has_stat)
school_crawl_logf($school_crawl_log, 7, 'Skipped some sections for : :');
school_crawl_logf($school_crawl_log, 7, "%s: %d", $reason, $num);
}
/*
* Calculate lab-based course dependencies.
*/
school_crawl_logf($school_crawl_log, 7, 'Adding implicit lab dependencies.');
foreach ($semester->departments_get() as $department)
foreach ($semester->department_classes_get($department) as $course)
{
$the_course = $semester->class_get($department, $course);
$lab_course = $semester->class_get($department, $course . 'L');
if (!empty($lab_course))
{
$the_course->dependency_add($lab_course);
school_crawl_logf($school_crawl_log, 8, "Adding dependency of %s-%s for %s-%s.",
$department, $course . 'L', $department, $course);
}
}
return 0;
}
/**
* \brief
* Add a course to a semester if that semester doesn't yet have this
* course.
*
* \param $semester
* The semester to which the course should be appended.
* \param $deparmtent
* The department of the course to add.
* \param $course_id
* The course_id which, with the department string, forms a
* fully-qualified course_id.
*/
function school_crawl_webadvisor_course_add(Semester $semester, $department, $course_id, $title)
{
if ($semester->class_get($department, $course_id) == NULL)
$semester->class_add(new Course($department . '-' . $course_id, $title));
}
/**
* \brief
* Find the datatelform and run it through school_crawl_form().
*
* \return
* See school_crawl_form(), NULL if form not found.
*/
function _school_crawl_webadvisor_form($dom, array &$school_crawl_log)
{
$xpath = new DOMXPath($dom);
foreach ($xpath->query('.//form[@name="datatelform"]') as $dom_form)
return school_crawl_form($dom_form);
school_crawl_logf($school_crawl_log, 2, "Unable to find form[@name=\"datatelform\"].");
return NULL;
}
/**
* \brief
* Returns the content of an element with the given ID.
*
* A convenience function.
*
* \param $domdocument
* A DOMDocument to search.
* \param $id
* The id attribute of the element whose content are requested.
* \return
* A UTF-8 string of the contents of the given element or NULL if
* the element isn't found.
*/
function _school_crawl_webadvisor_dom_id_content($domdocument, $id)
{
$node = $domdocument->getElementById($id);
if ($node)
{
return $node->nodeValue;
}
return NULL;
}