I need to perform a rather complicated chain of promise resolutions to get and save data that is located in PDFs that are uploaded by a user. Everything works fine for single PDFs, but it breaks as soon as I try to upload multiple PDFs because the order in which the resolution proceeds isn't as I expect.
I call the function saveDoc on each file in an array:
saveDoc: function(){
var files = this.$refs.upload.uploadFiles
var self=this;
var promises = []
for (var i=0; i<files.length; i++){
(function(){
var file = files[i]
var name = file['name']
if (!(/\.pdf/i.test(name))){
name+='.pdf'
}
var type = mime.lookup(name)
var file = file['raw'];
var beforeUrl = self.selectedCategories.join('&&&')
console.log('going to save: ' + name) // Because of the IIFE, I'm not expecting this to log only after all steps 1 - 20 have been completed for each file
Store.getFileData(file).then(function(data){
console.log(2)
return Store.saveDoc(name, type, data, beforeUrl).then(url => {
console.log(8)
return Text.getText(url).then(text => {
console.log(12)
return Text.getMetadata(text, url).then(metadata => {
console.log(20)
if (metadata.length){
return Store.saveMetadata(beforeUrl, metadata, name)
}
return Store.createCategory(name, self.selectedCategories, '')
})
})
})
})
})()
}
},
I'm sure the promises could use some work, but what seems to be the problem is that the line console.log('going to save: ' + name) is called twice before the entire sequence 1-20 is carried out for one file (see error message of number sequence at bottom of this post). I tried to prevent this using an IIFE, but I guess I didn't do this right.
store.js
getData: function(url){
return new Promise(function(accept, reject){
console.log(4)
documentation.get(url, function(err, body) {
console.log(5)
if (!err){
console.log(body);
accept(body)
}
});
})
},
getFileData: function(file){
var reader = new FileReader();
return new Promise(function(accept, reject){
console.log(1)
reader.onload = (e) => {
// var data = e.target.result.replace(/^data:[A-Za-z]+\/[A-Za-z]+;base64,/, '')
// console.log('base 64: ' + data)
var res = new Uint8Array(e.target.result)
accept(res)
};
reader.readAsArrayBuffer(file);
})
},
saveDoc: function(name, type, filedata, url){
console.log(3)
var self=this
return new Promise(function(accept, reject){
return self.getData(url).then(data => {
console.log(6)
var rev = data['_rev']
return documentation.attachment.insert(url, name, filedata, type,
{ rev: rev }, function(err, body) {
if (!err){
console.log(7)
var fullUrl = 'http://dev04/documentation/'+url+'/'+name
accept(fullUrl)
}
else {
console.log(err)
}
})
}).catch(err => {
console.log(err)
})
})
},
saveMetadata: function(url, metadata, name){
var fileName = path.basename(name)
var self=this
return new Promise(function(accept, reject){
self.getData(url).then(data => {
var meta
var rev = data['_rev']
if(!data['metadata']){
data['metadata'] = {}
}
data['metadata'][fileName] = metadata
var datastring = JSON.stringify(data)
documentation.insert(data, url, function(err, body, header) {
if (err) {
console.log(err.message);
return;
}
});
}).catch(err => {
console.log(err)
})
})
},
text.js
export default {
getText: function(url){
console.log(9)
var result = []
return new Promise(function(accept, reject){
console.log(10)
return pdfjs.getDocument(url).then(pdf => {
console.log(11)
var pdf = pdfjs.getDocument(url);
return pdf.then(function(pdf) { // get all pages text
var maxPages = pdf.pdfInfo.numPages;
var countPromises = []; // collecting all page promises
for (var j = 1; j <= maxPages; j++) {
var page = pdf.getPage(j);
var txt = "";
countPromises.push(page.then(function(page) { // add page promise
var textContent = page.getTextContent();
// console.log('the content is ' + textContent)
return textContent.then(function(text){ // return content promise
var val = text.items.map(function (s) { return s.str; }).join('&&&')+'&&&'; // value page text
result.push(val)
// console.log(val + ' should be one page of text')
return val
});
}));
}
// Wait for all pages and join text
return Promise.all(countPromises).then(function (texts) {
accept(texts.join(''))
});
});
});
})
},
getMetadata: function(text, url){
console.log(13)
var result = []
var self = this
return new Promise(function(accept, reject){
self.getpageno(url).then(function(pagecount){
console.log(19)
try {
var dataMatch = rx.exec(text)
var produktDaten = dataMatch[1].split("&&&").filter(Boolean);
console.log(produktDaten)
var dokuArr = dataMatch[2].split("&&&").filter(Boolean);
for (var i=0; i<produktDaten.length; i+=4){
var entry = {}
entry.pagecount = pagecount
entry.kks = {}
entry.kks.pages = {}
var kksNummer = produktDaten[i];
entry.kks.nummer = kksNummer;
for(var j=0; j<dokuArr.length; j+=3){
var nummer = dokuArr[j];
var beschreibung = dokuArr[j+1];
var seite = dokuArr[j+2];
// make sure seite is a digit
if (!(/^\d+$/.test(seite))){
console.log(seite + ' was not a valid page number')
throw err
}
if (/(A|a)lle?/i.test(nummer)){
entry.kks.pages[beschreibung] = seite;
// self.tableEntry.kks.url = url;
// self.tableEntry.fileName = name;
///// kksNummern.forEach(function(kks){
// self.tableEntry.kks;
// })
}
else if (nummer === kksNummer) {
entry.kks.pages[beschreibung] = seite;
// entry.kks.url = url;
// entry.fileName = name
}
}
entry.hersteller = produktDaten[i+1]
entry.typ = produktDaten[i+2]
entry.artikelNummer = produktDaten[i+3]
result.push(entry)
}
}
catch(e){
return accept(result)
}
return accept(result)
/* if (result.length>0){
console.log('accepting the result')
}
reject()*/
}).catch(err => {
console.log(err)
})
})
},
getpageno: function(url){
console.log(14)
var self=this
var pdf = pdfjs.getDocument(url);
return pdf.then(function(pdf){
console.log(15)
var maxPages = pdf.pdfInfo.numPages;
var countPromises = []; // collecting all page promises
for (var j = 1; j <= maxPages; j++) {
try {
var page = pdf.getPage(j);
var txt = "";
countPromises.push(page.then(function(page) { // add page promise
var textContent = page.getTextContent();
return textContent.then(function(text){ // return content promise
console.log(16)
return text.items.map(function (s) { return s.str; }).join('&&&'); // value page text
});
}));
}
catch(e){
console.log(e)
}
}
// Wait for all pages and join text
return Promise.all(countPromises).then(function (texts) {
// since doumentation pages do not add the final '&&&', must add one manually (only after rx has been found)
console.log(17)
var fulltext = texts.reduce(function(full, text){
if (rx.test(full)){
var next = '&&&'+text
return full+=next
}
return full+=text
}, '')
return [fulltext, texts]
});
}).then(function(textarr){
console.log(18)
var fulltext = textarr[0]
self.fulltext = fulltext;
var texts = textarr[1]
try {
var partialmatch = rx.exec(fulltext)[0]
var count = texts.reduce(function(pageno, text){
var tomatch = text.replace(/.*Typ&&&/, '')
if (tomatch.length>0 && partialmatch.indexOf(tomatch) > -1){
pageno++
}
return pageno;
}, 0)
}
catch(e){
console.log(e)
}
return count;
}).catch(err => {console.log(err)})
}
}
I use the console to log the order I am expecting for the functions. I am expecting the numbers 1 - 20 in order, but I get the following.
going to save: 03_.pdf selector.js:211:6
1 store.js:227:4
going to save: 2017.07.05_0016 E161206.pdf selector.js:211:6
1 store.js:227:4
2 selector.js:213:7
3 store.js:239:3
4 store.js:213:4
2 selector.js:213:7
3 store.js:239:3
4 store.js:213:4
5 store.js:215:5
Object { _id: "Test", _rev: "36-85a08c0852ccab78c0b4c10369e83fb2", rank: 7, icon: "wrench", metadata: Object, _attachments: Object } store.js:217:6
6 store.js:243:5
5 store.js:215:5
Object { _id: "Test", _rev: "36-85a08c0852ccab78c0b4c10369e83fb2", rank: 7, icon: "wrench", metadata: Object, _attachments: Object } store.js:217:6
6 store.js:243:5
7 store.js:247:7
8 selector.js:215:8
9 text.js:12:3
10
11 text.js:17:5
12 selector.js:217:9
13 text.js:50:3
14 text.js:112:3
15 text.js:116:4
16 text.js:129:8
17 text.js:142:5
18 text.js:153:4
19 text.js:57:5
20
Can anyone offer any advice on how to get this order correct? Thank you.
