Comments (42)
Note: that's the same <!--->
pattern in all of the reported files.
from jodd.
Here's a simple reproducer for the first page:
<html>
<body>
<!--->
-->
</body>
</html>
Please note the comment block is not closed in <!--->
as there's only 3 dashes (and not 4).
from jodd.
I want to upload all html pages that I used in the test. Unfortunately, it looks like Github has limit of 5GB per repository - I have 13GB. I will upload it to S3, but I need to return from travel - around 1 week.
I downloaded all the html and in http-client set by default UTF-8 encoding. In the specification I saw that finding a proper encoding is tricky. Can you recommend some library to do it?
from jodd.
@igr I run test aginst all these pages and there were no exceptions. I run the newest snapshot version
<repositories>
<repository>
<id>maven-snapshots</id>
<url>http://oss.sonatype.org/content/repositories/snapshots</url>
<layout>default</layout>
<releases>
<enabled>false</enabled>
</releases>
<snapshots>
<enabled>true</enabled>
</snapshots>
</repository>
</repositories>
<dependency>
<groupId>org.jodd</groupId>
<artifactId>jodd-lagarto</artifactId>
<version>6.0.0.20200731131821-SNAPSHOT</version>
<scope>test</scope>
</dependency>
Congratulations, you've fixed issue ;-)
from jodd.
I will migrate from Godday on the next payment cycle, too much for me atm...
@neroux makes sense, https://lagarto.jodd.org it is.
I will try to add Reader as an input, and then I will release lagarto to 6.
For now, I will update this repo and current docs.
from jodd.
No problem, I didnt forget this. In fact, I am moving Lagarto to a separate, individual project, as it deserves special attention - it is one of the two most used libraries of Jodd.
Thank you for helping with this:) Stay tuned, the plan is to migrate this week
from jodd.
These comments are going to kill me :)))))
from jodd.
I believe this one is fixed in last PR, but not released yet.
from jodd.
Are snapshots published on Sonatype or should I build locally?
from jodd.
Locally, sorry.
from jodd.
Will give it a try tomorrow then :)
from jodd.
Yap, just tested here, the above HTML snippet works on master
from jodd.
So hopefully you're fine now 🤞
from jodd.
Haha, this was close :)))
from jodd.
Hey @marekhudyma awesome! You can use https://wetransfer.com as well.
In a week - I promise - I will migrate to new repo and have the 6.0.0-snapshot ready!
Yeah, the encoding is tricky, but isn't it set by the response? Are you using Jodd http library? One idea is to download everything with same encoding and then resolve encoding from e.g. meta tags, and then reload string. Just na idea, thinking out loud here.
from jodd.
I found this definition of encodings and it is not trivial:
https://w3c.github.io/html-reference/syntax.html#character-encoding
That's why I ask for any automated way to find it. I will google more ;-)
I use Async Http Client, but I am still evaluating which async http client gives biggest configuration (for example stop downloading if file is bigger than x bytes).
I wrote some performance test that compare Jsoup, Jericho, HtmlCleaner, Lagarto with default configuration and Lagarto with tuned configuration. It looks Lagarto is the fastest ;-)
But I wanted to compare Jerry parser as well. Unfortunately documentation is not clear for me how to use it. I see for examples like:
Jerry doc = Jerry.jerry("some_html");
doc.$("div");
But my Intellij cannot resolve method "$". Is documentation not updated or I do something wrong?
from jodd.
I had to remove $
as in e.g. Graal it is not allowed. Try select
.
from jodd.
btw regarding speed ... its Lagarto < LagartoDOM < Jerry
from jodd.
Try also just s()
instead of $()
from jodd.
Here, new repo created: https://github.com/oblac/jodd-lagarto, migration is in progress :)
from jodd.
@marekhudyma see oblac/jodd-lagarto#3
I am thinking of way how to add a bunch of files for the tests. I guess I can put them in the Docker image so not to increase the repo size?
from jodd.
Thanks for this hint.
I need to admit the documentation is very outdated. Please fix it. Do you need an issue for it? ;-)
The LagartoDOM is so badly documented, I am not sure how to use it.
https://jodd.org/lagarto/dom-builder.html
It simply doesn't have this methods:
LagartoDOMBuilder domBuilder = new LagartoDOMBuilder() {
@Override
protected DOMBuilderTagVisitor createDOMDomBuilderTagVisitor() {
return new MyDOMBuilderTagVisitor();
}
}
from jodd.
I know @marekhudyma. I will put my efforts now in Lagarto. Even a separate web site :)
from jodd.
As I promised,
Here is a simple project with test of your parser. Test just check if there is no exception.. nothing complicated, but still provide some value as we discussed.
https://github.com/marekhudyma/html-parser-correctness
To make it work, you need to download 13 GB of downloaded pages. It is around 1 million (862_542) pages. Go to folder IN and execute aws S3 bucket synchronization:
aws s3 sync s3://html-top1m .
Then run: JoddLagartoCorrectnessTest
Parsing ~1m pages was not so long.
from jodd.
I got a problem to commit all files to GitHub, but maybe commiting 10 of them would be an option to have automated test. All of them you can run from time to time..
from jodd.
(I think there was a limit of 5GB for whole repository.. )
from jodd.
Just download the files :) I believe Docker image with all the files will make sense...
from jodd.
@marekhudyma what are your memory settings for the test? I have an issue with 27.zip
.
from jodd.
Soon... www.joddlagarto.com :)
from jodd.
@slandelle @marekhudyma @moh-sushi
Here is the snapshot that is a beta:
https://oss.sonatype.org/content/repositories/snapshots/org/jodd/jodd-lagarto/
6.0.0.20200731131821-SNAPSHOT
I made some minor API improvements, that is basically it.
The website is ready: www.joddlagarto.com :)
I will now remove Lagarto from here and do the big test, that is all whats left.
please, if you have some time, you can try this snapshot before its released :)
from jodd.
Would be great if LagartoDomBuilderConfig#setErrorLogger
and LagartoDomBuilderConfig#setDebugLogger
could return this
instead of void
so configuration could be chained.
Wondering about the perf impact of passing a Consumer<Supplier> though.
from jodd.
Ah true.
Regarding performance, I was thinking in skipping the construction of debug strings... but I will do that better now.
Thanx!
from jodd.
Nope :P
from jodd.
Yeah, I got carried away :)
from jodd.
@igr Nice website it comming.
I will run correctness test against snapshot using this project: https://github.com/marekhudyma/html-parser-correctness
I will be realistic: probably I will not build docker image from these files.. please handle it by yourself :)
I also write comparison test (between libraries). Probably it will takes around 2 weeks for me to return to this topic. The main concept is to compare with java-benchmarking - to show difference with 1 parsing, parsing 1million pages, to show who is faster in long run AND I want to compare DOM vs no-DOM parsers to show how many times you need to parse 1 document so building DOM makes sense.
from jodd.
Awesome @marekhudyma - you dont have to do anything, you already did enough :)
I am handling this locally, Docker images are also out of option, so it's going to be all locally.
Just please use the latest version, that is all that I ask :)
from jodd.
Soon... www.joddlagarto.com :)
Why not lagarto.jodd.org?
Also, why Godaddy? 😲 ;)
from jodd.
Why not lagarto.jodd.org?
Good question, @neroux... to be honest, I was looking for the quickest way to make documentation, and it seems that GitBook works so far. The new doc is totally separated, has a separate repository etc. - hence the new URL
I believe I can use a subdomain on jodd.org, but the Lagarto pages would look different from the rest of the existing website. Would that make sense?
I would probably then need to migrate the rest of the documentation too, to separate subdomains :)
Also, why Godaddy? 😲 ;)
10+ years ago it seemed like a good idea;)))
from jodd.
I believe I can use a subdomain on jodd.org, but the Lagarto pages would look different from the rest of the existing website. Would that make sense?
I would probably then need to migrate the rest of the documentation too, to separate subdomains :)
Please, no worries. I was just curious. Either approach will work just fine. Using hostnames of jodd.org
might just be cheaper in the end :)
10+ years ago it seemed like a good idea;)))
I know, I also was with that company once until their CEO went on hunting safaris for elephants and leopards.
from jodd.
I know, I also was with that company once until their CEO went on hunting safaris for elephants and leopards.
Damn, didn't know that. Do you have an alternative @neroux to recommend? I am bit tired of their non-working website anyway
from jodd.
Been using https://www.gandi.net for ages.
from jodd.
Recently I have been using mostly www.porkbun.com for gTLD domains.
from jodd.
Related Issues (20)
- optional return types for jodd-json HOT 3
- jodd-http set body encoding HOT 2
- Why LFUCache doesnot initialize with a loadFactor=1.0f of a Hashmap? HOT 14
- java.lang.IndexOutOfBoundsException HOT 8
- java.lang.IndexOutOfBoundsException HOT 17
- When I send one email with attachment, the attachment name is gibberish in Chinese. HOT 2
- envelopeOnly() results in lots of fetching compared to folder.search HOT 2
- jodd-petite singleton scope bean is thread safe?? HOT 1
- petite problem: when invoke addBean will throw MalformedParametersException HOT 4
- Can you integrate Swagger Doc like Spring frame? HOT 1
- Duplicate class HOT 22
- fix jodd-db document HOT 1
- InterfaceImplementation.java ClassCastException when interface return primitive type but not number HOT 1
- [How to] register custom action annotation? HOT 1
- Required OSGi headers missing HOT 3
- IDEA: Provide PathMacros in util module HOT 1
- LRUCache has Concurrency problem HOT 5
- tag missing problem
- 🧚🤖 Pixeebot Activity Dashboard
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from jodd.