We Keep Coding

Configurable HTML information extraction

Scenario:
I'm doing some HTML information extraction using a crawler. Right now, most the rules for extraction are hardcoded (not the tags or things like that, but loops, nested elements, etc.)
For instance, one common task is as follows:
Obtain table with ID X. If it doesn't exists there may be additional mechanisms so find the info which are triggered
Find a row which contains some info. Usually the match is a regexp against an specific column.
Retrieve the data in a different column (usually marked in the td, or previously detected in the header)
The way I'm currently doing so is:
Query to get the body of first table with id X (X is in config file). Some websites of my list are buggy and duplicate that id on elements different than table -.-
Iterate over interesting cells, executing regexp on cell.text() (regexp is in config file)
Get the parent row of the matching cells, and obtain the cell I need from the row (identifier of the row is in config file)
Having all this hardcoded for the most part (except column names, table ids, etc) gives me the benefit or being easy to implement and more efficiency than a generic parser, however, it is less configurable, and some changes in the target websites force me to deal with code, which makes it harder to delegate the task.
Question
Is there any language (preferably with a java implementation available) which allows to consistently define rules for extractions like those? I'm using css-style selectors for some tasks, but others are not so simple, so my best guess is that there must be something extending that that a non-programmer maintainer to add/modify rules on demand.
I would accept a Nutch-based answer, if there's one, as we're studying migrating our crawlers to nutch, although, I'd prefer a generic java solution.
I was thinking about writing a Parser generator and create my own set of rules to allow users/maintainers to generate parsers, but it really feels like reinventing the wheel for no reason.

I'm doing something somewhat similar - not exactly what you're searching for, but maybe you can get some ideas.
First the crawling part:
I'm using Scrapy on Python 3.7.
For my project, that brought the advantage, that it's very flexible and an easy crawling framework to build upon. Things like delays between requests, HTTP header language etc. can mostly be configured.
For the information extraction part and rules:
In my last generation of crawler (I'm now working on the 3rd gen, the 2nd one is still running but not as scalable) I've used JSON files to enter the XPath / CSS rules for every page. So on starting my crawler, I've loaded the JSON file for one specific page that is currently being crawled and a generic crawler, knew what to extract based on the loaded JSON file.
This approach isn't easily scalable since one config file per domain has to be created.
Currently, I'm still using Scrapy, with a starting list of 700 Domains to crawl and the crawler is now only responsible for downloading the whole website as HTML files.
These are being stored in tar archives by a shell script.
Afterward, a Python script is going through all members of the shell script and analyzing the content for the information I'm looking to extract.
Here, as you said, it's a bit like re-inventing the wheel or writing a wrapper around an existing library.
In Python, one can use BeautifulSoup for removing all tags like script and style etc.
Then you can extract for instance all text.
Or you'd focus first on tables only, extract all tables into dicts and can then analyze with regex or similar.
There are libraries like DragNet for boilerplate removal.
And there are some specific approaches on how to extract table structured information.

Reconciliation tool [comparing two large data set of records]

I have been asked to build a reconciliation tool which could compare two large datasets (We may assume input source as two excels).
Each row in excel contains 40-50 columns and record to be compared at each column level. Each file contains close to 3 million of records or roughly 4-5 GB of data.[data may not be in sorted format]
I would appreciate if i could get some hint.
Can following technologies be a good fit
Apache Spark
Apache Spark + Ignite [assuming real time reconciliation in between time frames]
Apache Ignite + Apache Hadoop
Any suggestion to build out in-house tool.

I have also been working on the same-
You can load the csv files to temporary tables using Pyspark/Scala and query on top of the temp tables created.

First a Warning:
Writing a reconciliation tool contains lots of small annoyances and edge cases like date formats, number formats (commas in numbers, scientific notation etc), compound keys, thresholds, ignoring columns , ignoring headers/footers etc etc.
If you only have one file to rec with well defined inputs then consider doing it yourself.
However, if you are likely to try to extend it to be more generic then pay for an existing solution if you can because it will be cheaper in the long run.
Potential Solution:
The difficulty with a distributed process is how you match the keys in unsorted files.
The issue with running it all in a single process is memory.
The approach I took for a commercial rec tool was to save the CSV to tables in h2 and use SQL to do the diff.
H2 is much faster than Oracle for something like this.
If your data is well structured you can take advantage of the ability of h2 to load directly from CSV and if you save the result in a table you can also write the output to CSV too or you can use other Frameworks to write a more structured output or stream the result to a web page.
If your format is xls(x) and not CSV you should do a performance test of the various libraries to read the file as there are huge differences when dealing with that size.

I have been working on the above problem and here is the solution.
https://github.com/tharun026/SparkDataReconciler
The prerequisites as of now are
Both datasets should have the same number of columns
For now, the solution accepts only PARQUETS.
The tool gives you match percentage for each column, so you could understand which transformation went wrong.

Java framework to manage BLOB data outside of database

I want to store my blobs outside of the database in files, however they are just random blobs of data and aren't directly linked to a file.
So for example I have a table called Data with the following columns:
id
name
comments
...
I can't just include a column called fileLink or something like that because the blob is just raw data. I do however want to store it outside of the database. I would love to create a file called 3.dat where 3 is the id number for that row entry. The only thing with this setup is that the main folder will quickly start to have a large number of files as the id is a flat folder structure and there will be OS file issues. And no the data is not grouped or structured, it's one massive list.
Is there a Java framework or library that will allow me to store and manage the blobs so that I can just do something like MyBlobAPI.saveBlob(id, data); and then do MyBlobAPI.getBlob(id) and so on? In other words something where all the File IO is handled for me?

Simply use an appropriate database which implements blobs as you described, and use JDBC. You really are not looking for another API but a specific implementation. It's up to the DB to take care of effective storing of blobs.

I think a home rolled solution will include something like a fileLink column in your table and your api will create files on the first save and then write that file on update.
I don't know of any code base that will do this for you. There are a bunch that provide an in memory file system for java. But it's only a few lines of code to write something that writes and reads java objects to a file.
You'll have to handle any file system limitations yourself. Though I doubt you'll ever burn through the limitations of modern file systems like btrfs or zfs. FAT32 is limited to 65K files per directory. But even last generation file systems support something on the order of 4 billion files per directory.
So by all means, write a class with two functions. One to serialize an object to a file; given it a unique key as a name. And another to deserialize the object by that key. If you are using a modern file system, you'll never run out of resources.

As far as I can tell there is no framework for this. The closest I could find was Hadoop's HDFS.
That being said the advice of just putting the BLOB's into the database as per the answers below is not always advisable. Sometimes it's good and sometimes it's not, it really depends on your situation. Here are a few links to such discussions:
Storing Images in DB - Yea or Nay?
https://softwareengineering.stackexchange.com/questions/150669/is-it-a-bad-practice-to-store-large-files-10-mb-in-a-database
I did find some addition really good links but I can't remember them offhand. There was one in particular on StackOverFlow but I can't find it. If you believe you know the link please add it in the comments so that I can confirm it's the right one.

Best way to compare two very large XML files record by record

I have two large XML files (3GB, 80000 records). One is updated version of another. I want to identify which records changed (were added/updated/deleted). There are some timestamps in the files, but I am not sure they can be trusted. Same with order of records within the files.
The files are too large to load into memory as XML (even one, never mind both).
The way I was thinking about it is to do some sort of parsing/indexing of content offset within the first file on record-level with in-memory map of IDs, then stream the second file and use random-access to compare those records that exist in both. This would probably take 2 or 3 passes but that's fine. But I cannot find easy library/approach that would let me do it. vtd-xml with VTDNavHuge looks interesting, but I cannot understand (from documentation) whether it supports random-access revisiting and loading of records based on pre-saved locations.
Java library/solution is preferred, but C# is acceptable too.

Just parse both documents simultaneously using SAX or StAX until you encounter a difference, then exit. It doesn't keep the document in memory. Any standard XML library will support S(t)AX. The only problem would be if you consider different order of elements to be insignificant...

Parsing very large XML files and marshalling to Java Objects

I have the following issue: I have very large XML files (like 300+ Megs), and I need to parse them in order to add some of their values to the db. The structure of these files is also very complex. I want to use Stax Parser as it offers the nice possibility of pull-parsing (and thus processing) only parts of the XML file at a time, and thus not loading the whole thing in memory, but on the other hand getting the values with Stax (at least on these XML files) is cumbersome, I need to write a ton of code. From this latter point of view it will immensly help me if I could marshall the XML file to Java objects (like JAX-B does) however this would load the whole file plus a ton of Object instances in memory all at once.
My question is, is there some way to pull-parse (or just partially parse) the file sequentially, and then marshall only those parts to Java objects so I can deal with them easily without bogging down on memory?

I would recommend Eclipse EMF. But it has the same problem, if you give it the file name it would parse the whole thing. Although there are some options to reduce how much is loaded, but I didn't bother much as we run on machines with 96 GB RAM. :)
Anyway, If your XML format is well defined, then one workaround is to fool the EMF by breaking down the whole file into several smaller (but still well defined) XML snippets. Then feed each snippet one after the other. I don't know JAX-B, but perhaps the same workaround can be applied there as well. Which I would recommend, because EMF is too big a hammer for such a small issue.
Just to elaborate a bit if your XML looks like this:
<tag1>
<tag2>
<tag3/>
<tag4>
<tag5/>
</tag4>
<tag6/>
<tag7/>
</tag2>
<tag2>
<tag3/>
<tag4>
<tag5/>
</tag4>
<tag6/>
<tag7/>
</tag2>
............
<tag2>
<tag3/>
<tag4>
<tag5/>
</tag4>
<tag6/>
<tag7/>
</tag2>
</tag1>
Then it can be broken down into one XML each starting with <tag2> and ending with </tag2>. And in java most parsers would accept a Stream, so just parse using whatever you want, create some StringStream or something for each <tag2> in a loop and pass to JAX-B or EMF.
HTH

Well, first off I wanna thank the two persons answering my questions, but I finally ended up not using those propositions partly because those proposed technologies are a bit far from the Java let's say "standard XML parsing" and it feels weird going so far when there's a similar tool already present in Java and partly also because in fact I did found a solution that only uses Java API's to accomplish this.
I will not detail too much the solution I found, because I've already finished the implementation, and it's quite a big chunk of code to place here (I use Spring Batch on top of it all, with a ton of configuration and stuff).
I will however make a small comment on what I finally ended up doing:
The big idea here is the fact that if you have an XML document AND it's corresponding XSD schema, you can parse & marshall it with JAXB, and you can do it in chunks, and said chunks can be read with an even parser such as STAX and then passed to the JAXB Marshaller.
This practically means that you must first decide where's a good place in your XML file where you can say "this part here has A LOT of repetive structure, I will treat those repetitions one at a time". Those repetitive parts are usually the same (child) tag repeated a lot inside a parent tag. So all you have to do is make an event listener in your STAX parser that is triggered at the start of each of those child tags, than stream over to JAXB the content of that child tag, marshall it with JAXB and process it.
Really the idea is excellently described in this article, which I followed (true, it's from 2006, but it deals with JDK 1.6 which at that time was pretty new, so version-wise it's not that old at all):
http://www.javarants.com/2006/04/30/simple-and-efficient-xml-parsing-using-jaxb-2-0/

Document projection might be the answer here. Saxon and a number of other XQuery processors offer this as an option. If you have a reasonably simple query that selects a small amount of data from a large document, the query processor analyses the query to work out which parts of the tree need to be available for the query, and which can be discarded during processing. The resulting tree can often be only 1% of the size of the full document. Details for Saxon here:
http://saxonica.com/documentation/sourcedocs/projection.xml