Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Updated Werelate scraper to use HTTPS #101

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Updated Werelate scraper to use HTTPS #101

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -122,5 +122,5 @@ from the same record on the same website.
* [Genealogie Online](https://www.genealogieonline.nl)
* [MyHeritage](https://www.myheritage.com) historical records and tree profiles
* [Open Archives](https://www.openarch.nl)
* [WeRelate](http://www.werelate.org/)
* [WeRelate](https://www.werelate.org/)
* [WikiTree](http://www.wikitree.com/) person profiles
94 changes: 47 additions & 47 deletions src/scrapers/werelate.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@ var debug = require('debug')('genscrape:scrapers:werela'),
GedcomX = require('gedcomx-js');

var urls = [
utils.urlPatternToRegex("http://www.werelate.org/wiki/Person:*")
utils.urlPatternToRegex("https://www.werelate.org/wiki/Person:*")
];

module.exports = function(register){
Expand Down Expand Up @@ -86,9 +86,9 @@ var factTypes = {
};

function run(emitter){

debug('run');

var gedx = GedcomX(),
primaryPerson = GedcomX.Person({
principal: true,
Expand All @@ -97,13 +97,13 @@ function run(emitter){
'genscrape': getRecordIdentifier(document.location.href)
}
});

gedx.addPerson(primaryPerson);

//
// Facts
//

// Gather the fact data
var facts = [];
Array.from(document.querySelectorAll('.wr-infotable-factsevents tr')).forEach(function(row){
Expand All @@ -115,21 +115,21 @@ function run(emitter){
});
}
});

// Process the fact data
facts.forEach(function(factInfo){

var row = factInfo.row,
label = factInfo.label,
dateCell = row.children[1],
placeCell = row.children[2];

switch(label){

case 'Name':
primaryPerson.addSimpleName(dateCell.textContent.trim());
break;

case 'Gender':
switch(dateCell.textContent.trim()){
case 'Male':
Expand All @@ -144,7 +144,7 @@ function run(emitter){
break;
}
break;

// Most facts will have a date and a place
default:
if(row.children.length === 3){
Expand Down Expand Up @@ -173,30 +173,30 @@ function run(emitter){
}
}
}

});

//
// Relationships
//

// Parents and siblings
Array.from(document.querySelectorAll('.wr-infobox-parentssiblings')).forEach(function(family){

var parents = family.querySelector('ul'),
children = family.querySelector('ol'),
marriage = family.querySelector('.wr-infobox-event'),
parent1Label = parentLabel(parents.children[0]),
mother, father;

if(parent1Label === 'F'){
father = processPerson(parents.children[0], 'http://gedcomx.org/Male');
mother = processPerson(parents.children[1], 'http://gedcomx.org/Female');
} else {
father = processPerson(parents.children[1], 'http://gedcomx.org/Male');
mother = processPerson(parents.children[0], 'http://gedcomx.org/Female');
}

if(father){
gedx.addPerson(father);
gedx.addRelationship({
Expand All @@ -205,7 +205,7 @@ function run(emitter){
person2: primaryPerson
});
}

if(mother){
gedx.addPerson(mother);
gedx.addRelationship({
Expand All @@ -214,7 +214,7 @@ function run(emitter){
person2: primaryPerson
});
}

if(father && mother){
var couple = GedcomX.Relationship({
type: 'http://gedcomx.org/Couple',
Expand All @@ -231,25 +231,25 @@ function run(emitter){
}
gedx.addRelationship(couple);
}

Array.from(children.children).forEach(function(child){

// Skip the entry for the primary person
if(child.querySelector('.selflink')){
return;
}

child = processPerson(child);
gedx.addPerson(child);

if(father){
gedx.addRelationship({
type: 'http://gedcomx.org/ParentChild',
person1: father,
person2: child
});
}

if(mother){
gedx.addRelationship({
type: 'http://gedcomx.org/ParentChild',
Expand All @@ -259,17 +259,17 @@ function run(emitter){
}
});
});

// Spouses and children
Array.from(document.querySelectorAll('.wr-infobox-spousechildren')).forEach(function(family){

var parents = family.querySelector('ul'),
children = family.querySelector('ol'),
marriage = family.querySelector('.wr-infobox-event'),
parent1Label = parentLabel(parents.children[0]),
parent2Label = parentLabel(parents.children[1]),
spouse, spouseLabel, couple;

// Determine whether the primary person is the husband or the wife
if(parents.children[0].querySelector('.selflink')){
spouse = processPerson(parents.children[1]);
Expand All @@ -278,19 +278,19 @@ function run(emitter){
spouse = processPerson(parents.children[0]);
spouseLabel = parent1Label;
}

spouse.setGender({
type: spouseLabel === 'H' ? 'http://gedcomx.org/Male' : 'http://gedcomx.org/Female'
});

gedx.addPerson(spouse);

couple = GedcomX.Relationship({
type: 'http://gedcomx.org/Couple',
person1: primaryPerson,
person2: spouse
});

if(marriage){
couple.addFact({
type: 'http://gedcomx.org/Marriage',
Expand All @@ -299,21 +299,21 @@ function run(emitter){
}
});
}

gedx.addRelationship(couple);

if(children){
Array.from(children.children).forEach(function(child){

child = processPerson(child);
gedx.addPerson(child);

gedx.addRelationship({
type: 'http://gedcomx.org/ParentChild',
person1: primaryPerson,
person2: child
});

gedx.addRelationship({
type: 'http://gedcomx.org/ParentChild',
person1: spouse,
Expand All @@ -322,9 +322,9 @@ function run(emitter){
});
}
});

// TODO: gather sources listed in the profile

// Agent
gedx.addAgent(GedcomX.Agent()
.setId('agent')
Expand All @@ -333,9 +333,9 @@ function run(emitter){
value: 'WeRelate'
})
.setHomepage({
resource: 'http://www.werelate.org'
resource: 'https://www.werelate.org'
}));

// SourceDescription
gedx.addSourceDescriptionToAll({
about: document.location.href,
Expand All @@ -346,22 +346,22 @@ function run(emitter){
],
citations: [
{
value: document.title + ', WeRelate.org (' + window.document.location.href
value: document.title + ', WeRelate.org (' + window.document.location.href
+ ' : accessed ' + utils.getDateString() + ')'
}
],
repository: {
resource: '#agent'
}
});

emitter.emit('data', gedx);
}

/**
* Get the label for a parent in a family box. For parents it is F (Father) or
* M (Mother). For spouses it's H (Husband) or W (Wife).
*
*
* @param {Element} parent Parent's li element
* @return {String} label
*/
Expand All @@ -373,7 +373,7 @@ function parentLabel(parent){

/**
* Create a GedcomX.Person from a person entry in a family box
*
*
* @param {Element} element Person's li DOM element
* @param {String} gender
* @return {GedcomX.Person}
Expand Down Expand Up @@ -417,7 +417,7 @@ function processPerson(element, gender){

/**
* Get the record ID
*
*
* @param {String} url
* @return {String}
*/
Expand All @@ -427,7 +427,7 @@ function getRecordId(url) {

/**
* Get a record's identifier
*
*
* @param {String} url
* @return {String}
*/
Expand Down
6 changes: 3 additions & 3 deletions test/data/werelate/output/mary-ball.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -730,10 +730,10 @@
"id": "1",
"citations": [
{
"value": "Person:Mary Ball (5) - Genealogy, WeRelate.org (http://www.werelate.org/wiki/Person:Mary_Ball_%285%29 : accessed 17 April 2013)"
"value": "Person:Mary Ball (5) - Genealogy, WeRelate.org (https://www.werelate.org/wiki/Person:Mary_Ball_%285%29 : accessed 17 April 2013)"
}
],
"about": "http://www.werelate.org/wiki/Person:Mary_Ball_%285%29",
"about": "https://www.werelate.org/wiki/Person:Mary_Ball_%285%29",
"titles": [
{
"value": "Person:Mary Ball (5) - Genealogy"
Expand All @@ -754,7 +754,7 @@
}
],
"homepage": {
"resource": "http://www.werelate.org"
"resource": "https://www.werelate.org"
}
}
]
Expand Down
6 changes: 3 additions & 3 deletions test/data/werelate/output/washington.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -713,10 +713,10 @@
"id": "1",
"citations": [
{
"value": "Person:George Washington (6) - Genealogy, WeRelate.org (http://www.werelate.org/wiki/Person:George_Washington_%286%29 : accessed 17 April 2013)"
"value": "Person:George Washington (6) - Genealogy, WeRelate.org (https://www.werelate.org/wiki/Person:George_Washington_%286%29 : accessed 17 April 2013)"
}
],
"about": "http://www.werelate.org/wiki/Person:George_Washington_%286%29",
"about": "https://www.werelate.org/wiki/Person:George_Washington_%286%29",
"titles": [
{
"value": "Person:George Washington (6) - Genealogy"
Expand All @@ -737,7 +737,7 @@
}
],
"homepage": {
"resource": "http://www.werelate.org"
"resource": "https://www.werelate.org"
}
}
]
Expand Down
12 changes: 6 additions & 6 deletions test/src/scrapers/werelate.js
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,15 +1,15 @@
var setupTest = require('../../testHelpers').createTestRunner('werelate');

describe('werelate', function(){

it('male', setupTest(
'washington',
'http://www.werelate.org/wiki/Person:George_Washington_%286%29'
'https://www.werelate.org/wiki/Person:George_Washington_%286%29'
));

it('female and children', setupTest(
'mary-ball',
'http://www.werelate.org/wiki/Person:Mary_Ball_%285%29'
'https://www.werelate.org/wiki/Person:Mary_Ball_%285%29'
));

});