I am evaluating feathers.js for a project. I like its aspirations. So, I decided to try and build a basic content management system just as a learning endeavor. Things have gone pretty smoothly. However, I want to load some static content (articles) into memory when the app starts. I can't figure out how to do that.
I have my articles in the data/articles directory. Each article is markdown named [title].md. I have a block of JavaScript that I tested in a console app that converts the markdown to HTML. That code uses markdown-js to get the HTML into a JSON object. It looks like this:
const fs = require('fs');
const markdownConverter = require('markdown');
let articles = [];
let files = fs.readdirSync('./data/articles');
for (let i=0; i<files.length; i++) {
let title = files[i].substr((files[i].lastIndexOf('/')+1), (files[i].length-3));
let markdown = fs.readFileSync(files[i], 'utf8');
let html = markdownConverter.toHTML(markdown);
articles[title] = html;
}
I've added a route in Feathers that works like this:
app.use('/articles/:slug', function(req, res) {
console.log('loading article: ' + req.params.slug);
let content = '';
// TODO: How to get access to the articles array.
// I want to get the HTML using content = articles[req.params.slug];
res.render('article', { content: content });
});
I'm not sure where to put the code that loads the markdown into an array that I can access when a user requests an article. Where does that belong? My guess is in the app.js file that gets generated when you create a Feathers project using the yeoman generator. Yet, I'm not sure what that actually looks like.
Since feathers is an Express app, you should be able to use express middleware. I recommend this one that allows you to create HTML templates for you markdown and serve them statically without creating any parsers or for loops.
https://github.com/natesilva/node-docserver
var docserver = require('docserver');
...
app.use(docserver({
dir: __dirname + '/docs', // serve Markdown files in the docs directory...
url: '/'} // ...and serve them at the root of the site
));
Or, this middleware that will preParse the Markdown before serving it as HTML in a Jade template.
https://www.npmjs.com/package/markdown-serve
Related
I have a personal website written in AngularJS and NodeJS, and in the tech projects section I would like to embed README.md from a project on GitHub my account (the same way github.com) does it. I don't think iframe<> works for md files. So I have tried converting to HTML and PDF.
The second part of the README.md I want looks like this:
In particular, I would like to import the README.md files into my HTML. I have found two ways to do it:
Download the file, and convert to HTML.
var md_to_html = function(src, dest) {
// Get the markdown
var md = fs.readFileSync(src, "utf8");
var converter = new Remarkable();
var html = converter.render(md);
fs.writeFile(dest, html, function(err) {
if (err) {
return console.log(err);
}
console.log(dest + " was saved!");
});
}
Or Download the file, and convert to PDF.
var md_to_pdf = function(src, dest) {
fs.createReadStream(src)
.pipe(markdownpdf())
.pipe(fs.createWriteStream(dest));
}
Both work, however the photo disappears. That is because the dependencies let's say doc/img/photo.jpg does not get downloaded with the README. I have also written scripts to download the dependencies. But neither insert the photo into the README.pdf or README.html.
Has anyone done something like this? Embedded markdown in HTML with photos, or converted them to PDF or HTML with photos? If I could just import it kind of like a PDF with iframe<> that would be ideal, but if not, I will program it myself. But how?
How to create PDF Documents in Node.JS.? Is there any better solution to manage templates for different types of PDF creation.
I am using PDFKit to create PDF Documents and this will be server side using Javascript. I can not use HTML to create PDF. It will blob of paragraphs and sections with replacing tags with in.
Does anyone know Node.js has any npm package that can deal templates with paragraphs sections headers.
Something like
getTemplateByID() returns a template that contains sections , headers, paragraphs and then i use to replace appropriate tags within the template.
In my case, I have to get my HTML template from my database (PostgreSQL) stocked as stream. I request the db to get my template and I create a tmp file.
Inside my template, I have AngularJS tags so I compile this template with datas thanks to the 'ng-node-compile' module:
var ngCompile = require('ng-node-compile');
var ngEnvironment = new ngCompile();
var templateHTML = getTemplateById(id);
templateHTML = ngEnvironment.$compile(templateHTML)(datas);
Now I have my compiled template (where you can set your paragraph etc.) and I convert them into PDF thanks to a PhantomJS module 'phantom-html-to-pdf'
var phantomHTML2PDF = require('phantom-html-to-pdf')(options);
phantomHTML2PDF(convertOptions, function (error, pdf) {
if(error) console.log(error);
// Here you have 'pdf.stream.path' which is your tmp PDF file
callback(pdf);
});
Now you have your compiled and converted template (pdf), you can do whatever you want ! :)
Useful links:
https://github.com/MoLow/ng-node-compile
https://github.com/pofider/phantom-html-to-pdf
I hope this help !
I need to parse json object when the node server.js(which is my entry point to the program) is started ,the parse of the json file is done in diffrent module in my project.
I've two questions
Is it recommended to invoke the parse function with event in the server.js file
I read about the event.emiter but not sure how to invoke function
from different module...example will be very helpful
I've multiple JSON files
UPDATE to make it more clear
if I read 3 json file object (50 lines each) when the server/app is loaded (server.js file) this will be fast I guess. my scenario is that the list of the valid path's for the express call is in this json files
app.get('/run1', function (req, res) {
res.send('Hello World!');
});
So run1 should be defined in the json file(like white list of path's) if user put run2 which I not defined I need to provide error so I think that when the server is up to do this call and keep this obj with all config valid path and when user make a call just get this object which alreay parsed (when the server loaded ) and verify if its OK, I think its better approach instead doing this on call
UPDATE 2
I'll try explain more simple.
Lets assume that you have white list of path which you should listen,
like run1
app.get('/run1', function
Those path list are defined in jsons files inside your project under specific folder,before every call to your application via express you should verify that this path that was requested is in the path list of json. this is given. now how to do it.
Currently I've develop module which seek the json files in this and find if specific path is exist there.
Now I think that right solution is that when the node application is started to invoke this functionality and keep the list of valid paths in some object which I can access very easy during the user call and check if path there.
my question is how to provide some event to the validator module when the node app(Server.js) is up to provide this object.
If it's a part of your application initialization, then you could read and parse this JSON file synchronously, using either fs.readFileSync and JSON.parse, or require:
var config = require('path/to/my/config.json');
Just make sure that the module handling this JSON loading is required in your application root before app.listen call.
In this case JSON data will be loaded and parsed by the time you server will start, and there will be no need to trouble yourself with callbacks or event emitters.
I can't see any benefits of loading your initial config asynchronously for two reasons:
The bottleneck of JSON parsing is the parser itself, but since it's synchronous, you won't gain anything here. So, the only part you'll be able to optimize is interactions with your file system (i.e. reading data from disk).
Your application won't be able to work properly until this data will be loaded.
Update
If for some reason you can't make your initialization synchronous, you could delay starting your application until initialization is done.
The easiest solution here is to move app.listen part inside of initialization callback:
// initialization.js
var glob = require('glob')
var path = require('path')
module.exports = function initialization (done) {
var data = {}
glob('./config/*.json', function (err, files) {
if (err) throw err
files.forEach(function (file) {
var filename = path.basename(file)
data[filename] = require(file)
})
done(data);
})
}
// server.js
var initialization = require('./initialization')
var app = require('express')()
initialization(function (data) {
app.use(require('./my-middleware')(data))
app.listen(8000)
})
An alternative solution is to use simple event emitter to signal that your data is ready:
// config.js
var glob = require('glob')
var path = require('path')
var events = require('events')
var obj = new events.EventEmitter()
obj.data = {}
glob('./config/*.json', function (err, files) {
if (err) throw err
files.forEach(function (file) {
var filename = path.basename(file)
obj.data[filename] = require(file)
})
obj.emit('ready')
})
module.exports = obj
// server.js
var config = require('./config')
var app = require('express')()
app.use(require('./my-middleware'))
config.on('ready', function () {
app.listen(8000)
})
I am having trouble writing to a file in Titanium Studio.
specifically .json file. Code is compiled through and no exception was thrown.
Here is my relevant section of code, I parse the file to var first before adding element and stringify it to be written back.
Reading works perfectly, so is adding element, it's the writing process that has issues
var file = Ti.Filesystem.getFile(Ti.Filesystem.resourcesDirectory,'data.json');
var jsontext = file.read().toString();
var jsondoc = JSON.parse(jsontext);
jsondoc['feedlist'].push({
"picloc":imagename,
"title":titlef.value,
"desc1":descf1.value,
"desc2":descf2.value,
"desc3":descf3.value
});
jsontext = JSON.stringify(jsondoc);
file.write(jsontext); // write(data,[append])
Note: I have consulted Documentation and done some of my own search, some are suggesting that "Filestream" should be used in place of normal file along with .close(), I have yet got them working but it could be pointers the solution, if anyone knows how to get it working
Thanks in advance.
EDIT: This question is flagged for duplication, initially I deemed that was 2 separate issues, one was about merely writing text to a file. Another is parsing event.media (picture) into a file.
I got it working now, The issue was that I was trying to write to file in read-only directory
Ti.Filesystem.resourcesDirectory: A read-only directory where your application resources are located
Ti.Filesystem.applicationDataDirectory: A read/write directory accessible by your app. Place your application-specific files in this directory.
The contents of this directory persist
until you remove the files or until the user uninstalls the application
Here is my code, directory is modified
var sesfile = Ti.Filesystem.getFile(Ti.Filesystem.applicationDataDirectory,'data2.json');
var jsontext = sesfile.read().toString();
var jsondoc = JSON.parse(jsontext);
jsondoc['feedlist'].push({
"picloc":imagename,
"title":titlef.value,
"desc1":descf1.value,
"desc2":descf2.value,
"desc3":descf3.value
});
jsontext = JSON.stringify(jsondoc);
sesfile.write(jsontext,false);
If you are unable to locate data directory and simply want to load the file from there.
(In my case it does not exist in project nor will be created with Webpreview compilings)
You can do bootstrap-ish type instruction like this first
var rdfile = Ti.Filesystem.getFile(Ti.Filesystem.resourcesDirectory,'data.json');
var sesfile = Ti.Filesystem.getFile(Ti.Filesystem.applicationDataDirectory,'data2.json');
var jsontext = rdfile.read().toString();
var jsondoc = JSON.parse(jsontext);
sesfile.write(jsontext);
hope it helps whomever makes amateur mistake like I did.
Basically, I wrote a server that response a js file(object format) to users who made the request. The js file is generated by two config file. I call them config1.js and config2.js.
Here is my code:
var express = require('express');
var app = express();
var _ = require('underscore');
app.use('/config.js', function (req, res) {
var config1 = require('config1');
var config2 = require('config2');
var config = _.extend(config1, config2);
res.set({'Content-Type': 'application/javascript'});
res.send(JSON.stringify(config));
});
For what I am understanding, every time I make a request to /config.js, it will fetch the latest code in config1 and config2 file even after I start server. However, if I start server, make some modification in config1.js. then make the request, it will still return me the old code. Can anyone explain and how to fix that? thanks
You should not use require in order to load your files because it is not its purpose, it caches the loaded file (see this post for more information), that is why you get the same content every time you make a request.
Use a tool like concat-files instead, or concat it "by hand" if you prefer.
Concat files and extend objects aren't equal operations. You can read the files via 'fs' module, parse objects, extend, and send.