I'm writing an embedded C++ program, and need to add serialization/deserialization. The format should be human readable and writeable, and I would much prefer to use (a subset of) a standard format like YAML. I also prefer YAML to JSON since it is more concise.
While yaml-cpp has the exact functionality I'd like, the source code is almost 300K and would almost double my code size, which seems excessive to me just in order to add human readable serialization/deserialization.
Before I start writing my own reader/writer for a subset of YAML, I'd like to first check whether this already exists? I have not been able to find one, but would much prefer to use existing code rather than rolling my own. Are there any C or C++ YAML readers/writers out there of, say, 50K code or less? I only need functionality for the basic data structures (scalar, array, hash), not any advanced stuff.
With many thanks in advance.
The Oops library is doing what you are looking for. It is written for serialization using reflection and supports YAML format as well.
https://bitbucket.org/barczpe/oops
Related
I have a perl script which generates a very large data structure (which starts life as an array of array references). This is then written to a text file using some weird home-brew serialisation scheme.
The data from the text file is stored as the value in a key-value store db.
A c++ file then retrieves the data, and deserializes it (into a hashmap, although can potentially be flexible on how this data is structured).
What I'm interested in is finding if there are any good ways of sharing a data structure between perl and c++ (something like Storable, but that is meant for perl->perl not perl->c++). The current method is a headache to maintain, and may not have the best performance.
The most important factors are speed of deserialisation, and the size of the serialized structure in that order. Anyone know of something that might do the trick?
Storable is one way to dump and load perl data structures. I wouldn't actually recommend it for general usage though - it's handy in that it's part of core and easy to use.
But for multi-platform (and language) portability, it's far better to use a standard data representation. Which you choose is probably a matter of what sort of data you're holding in your structure, but core contenders are:
JSON - good for arrays and hashes (key-value).
YAML - Excellent for 'config file' style data (but extends in ways similar to JSON)
And if you must, XML - but bear in mind that XML is designed for documents-with-metadata, and so IMO isn't suitable for most of the applications it's used for.
As standards, they've got documented formatting and parsers are widely available. And implementing your own isn't too hard, if that's the route you want to go. Just make sure you follow the spec and you're good.
Note - that because XML and JSON (and I think YAML?) are recursive, you can parse as a stream, rather than a standalone object. (Trap, process and discard as you hit 'close brackets' in JSON, or 'close tags' in XML).
easy job.
I like perl , and I also like C/C++. To make the best of both,
I wrote a github project to solve this issue.
please see:
https://github.com/tlqtangok/perlcpp
a short example is here :
P_eval("$a=2;$a=$a**10;");
Int("a") ; // a= 1024
I need to model a JSon in C++. Firstly I thought about:
boost::property_tree
But unfortunately - it does not care about JSon types. Integers are represented as strings - and it is not an option.
The second idea would be:
boost::variant with recursive_wrapper
Which looks quite promising.
What kind of approach could you recommend? Do you know better approaches? It looks like a common problem, so there have to be a lot of well tested solutions.
I'm unfamiliar with C++ JSon libraries. If you could recommend any - I would be grateful as well.
For JSon you typically want to take a library. You could ofcourse write your own implementation (parser etc.) but I don't see why you would do that considering there are so many good ones freely available.
Some examples:
https://github.com/kazuho/picojson (Header only)
https://github.com/miloyip/rapidjson
I have a large file of source code that I need to parse some specific text out of. I want to get it done as fast as possible. What would be the fastest way to do this in Swift? These are all the options I could think of?
Using a third-party library of string functions- I've tried this. It works well, but I imagine this is much slower compared to other lower level methods in general, unless there are some notably fast ones out there specifically for Swift.
Using a third-party HTML parser. I've looked into a few, but I'm not sure if they will suit my needs. Before I proceed with this, I just want to know if these are generally faster, if there are any notabley fast ones out there, and if I'm able to tweak them to get specifically what I want from the source code.
Using String or NSString. From what I understand, using String vs NSString should give no difference in speed. I am quite comfortable with this approach, and it's lower level than some of the other ones, so should I expect fairly fast performance?
Using regular expressions. I've been told that since these are lower-level, they should ideally be the fastest. I've used regular expressions before, but not in ios. Is it easy to do string parsing with NSRegularExpression, and is it faster?
Thank you!
Came upon this link while researching your question: http://benedictcohen.co.uk/blog/archives/74
The authors explains an older approach to what #CodaFi suggested, but there is a relevant update at the end you should check out:
The easiest way to parse HTML is to treat it as XML and use the
NSXMLParser. iOS comes with LibTidy which is capable of fixing a
multitude of markup sins. Use LibTidy to create clean XML and pass
this XML to NSXMLParser. Only use the approach outlined above if it’s
not possible to use NSXMLParser.
So perhaps option 4 or 5 for you to check out?
This question already has answers here:
Is there a C++ library to read JSON documents into C++ objects? [duplicate]
(4 answers)
Closed 8 years ago.
I can't figure out whether or not to use XML, YAML, or JSON for a C++ 2D RPG.
Here are my thoughts:
I need something which is simple to save not just player data, but environment data, such as object (x, y) coordinates; load times; dates; graphics configurations, etc.
I need something flexible, easy to use, and definitely light weight, but powerful to handle the above.
Which is the best choice? I have experience with JSON in JavaScript, but not C++. Are there any good references for parsing JSON in C++ if this is the route to go?
Edit
Honestly, if a text file seems like the simplest and most effective solution for something like this (especially if I can just write it to binary), then I'm all ears.
Edit 2
Feel free to provide other suggestions as well.
I would use the simplest thing that satisfies your requirements.
If you don't need hierarchical storage, then flat tabular files are so much easier to deal with than anything else. All you have to do is read lines off disk and split on tab.
If you are looking at more of key/value pair type storage (as opposed to lists of things), then INI files can be reasonable. This format has a lot of flexibility, though reasoning about it can less approachable when you start doing more complicated things than it was designed for.
If you need hierarchical, it's possible that JSON would be simpler. There are JSON libraries in wide range of languages, and it sounds like you already familiar.
https://stackoverflow.com/questions/245973/whats-the-best-c-json-parser
sqlite may be another option. There be dragons in SQL, but with a nice C++ wrapper around sqlite, it can be manageable. The primary benefit would be ACID, in my opinion.
The YAML spec looks somewhat lengthy, so I can guess that it has more kitchen sinks. Just skimming the libyaml docs, the API looks somewhat like SAX interfaces that I've used in the past. I have no a posteriori knowledge of it, but I would be reticent to start using it without a good reason.
XML sucks to deal with, don't opt in to it. There lots of reasons for this. I think the most relevant one in my mind is that it's prone to make the code that uses it more complicated than it would be otherwise. Any system I've seen designed with XML, reasoning about the XML is more complicated than the design interests that its trying to support. There are valid uses for it, though it's rare that another storage system wouldn't have been just as adequate.
Regardless of which one you choose, write as little code as you can managing it. You really want to write the classes your engine will use first. Then worry about serializing them. If you let your serialization influence your class design, you'll probably regret it. :)
I'm working on a C++ project, and wanted to get some inputs from developers with similar experience.
The task is to connect to a web service which gives the results in an XML form. My role in the task is once I receive the XML form, I need to convert the XML into a C++ object and parse the XML data to the C++ object.
Following are my clarifications.
a) One way is to handcraft the whole thing but I need to do this for around hundreds of web services. I am aware there are simpler tools for C# and Java to do the same.
Is there a tool/utility for C++ too?
Any suggestions, would be helpful.
In the past, I've used TinyXML for my XML parsing needs. My parsing code operated under the assumption that all XML input conforms to a particular XSD schema I wrote. It worked fairly well but the ripple effects were annoying - if I wanted to change the XSD, I had to update all my XML test files as well as my parsing code. While it's not so bad in the case of parsing one schema, I'd hate to have to do it for hundreds of them.
I'm not sure what the common solution is, but CodeSynthesis XSD sounds pretty promising. I haven't used it, but it appears that it generates a data layer, a parser and serialisation code for you. Could save you a lot of time.
If you're asking if there's a way to dynamically create an object representation of an XML data stream (such that you can access it like topLevel.subObject.value), it's not possible. C++ is a statically-typed language, which means all objects need to be defined a compile time. The best you could do is something like: xmlData.getSubObject("objectName").getValue().
As for toolsets for parsing into something usable dynamically (as per my later example), there are several. For Windows, for example, you could use the "built-in" MSXML objects. There's nothing in the base C++ libraries to do so, however, as far as I am aware.
Hope that helps.