Has anyone here tried it or is it possible?
I've been using PHP for quite a few years but never know exactly the underlying c scripts.
Is there a way to go into it?
I've done a bit of hacking on Zend PHP. I find it to be overly clever , some people go as far as calling it deliberately obfuscated in plain view. The source code to PHP is a mind altering (or breaking) substance, depending on how good you are at deciphering very cryptic macros. That's my impression of the core.
Writing extensions, however, is a breeze once you get the hang of the Zend helpers, most people with advanced-beginner / intermediate knowledge of C could get through a basic extension. There's also plenty of examples. One of the best parts of PHP is how organized the build system is, dropping in new stuff is relatively painless. With a little work and patience, almost any C library is rather easily extended to PHP.
If you aren't well versed in C (and what borders on abuse of the preprocessor), a glance at the PHP core is not going to give you much insight, nor is it a good thing to reference if you are learning C on your own.
Moving On:
Don't let anything I've said, or what anyone else has to say discourage you from grabbing the code and looking for yourself. That being said, as for debugging goes:
Valgrind (unless you use a lot of suppressions) is not very helpful. PHP (to the best of my knowledge) uses arch optimized reads, similar to new versions of glibc. I.e. its going to read 32 bits even if it only swallows 8 bits and a trailing NULL.
I have never found GDB to be very helpful with PHP. A lot of the magic is in macros that are very hard to trace.
You will quickly see the Zend error logging functions, and their version of assertions. Use those, printf() debugging is pretty much useless unless your debugging a CLI app.
Garbage collection can make you see odd things when using tools like valgrind's massif. Profiling heap use in PHP is an art I have not yet discovered.
Finally, I'd like to say its always nice to see someone take a look under the hood of their language. SO could use some questions that helps de-mystify the PHP core, so please feel free to post more as you go :)
Also, remember, Zend isn't the only forge that makes php. While compatibility with Zend is paramount if you hope for it to be adopted, everyone is still free to do their own thing.
I've never debugged the C code of PHP (nor extensions), but I've sometimes generated backtraces, in cases of crashes in PHP extensions.
This page might help, about that : Generating a gdb backtrace.
Starting from there, maybe you'll be able to go farther...
From some of your questions it sounds like you don't know that PHP is open-source, you can download the complete source code of it and look through all the C functions. If you want backtracing and debugging ability, you have to do what Pascal MARTIN said.
Related
I would like to be able to write database driven applications (i.e. standalone apps that are not web based and don't require a browser or Apache server to run)
I have attempted to do this in Codegear C++ Builder in the past but even though my 'background' is in that (C++ OOP with Borland Builder) it is so far removed from doing the same sort of thing with PHP/mysql and other web technologies that I found that I couldn't get very far for a lot of effort getting it to work. It was a while ago now but I was using the built-in database engine that comes with Builder and I just found it frustrating and difficult.
In other words - Is there something out there that will allow me to make use of web based languages to write standalone applications (specifically PHP/Javascript/mysql)
You can stick with PHP if you want. There are QT bindings, GTK bindings, OSX/Cocoa bindings, and you can call Win32 functions. I don't know how stable all of those are, but you can do GUI in PHP as well as the command line stuff.
As for other languages... PHP is very C like. It started as basically a scripting wrapper around C (IIRC), which is why you have functions named after the C standard library (like strstr). C like languages will feel quite familiar.
I would think Python would probably be the closest to PHP. It's a scripting language, syntax is somewhat close, it has a ton of libraries, and is very well supported and commonly used. I'd imagine it would feel pretty familiar. Using indenting instead of brackets for blocks can throw some people, but it fit the way I already indented my code.
Ruby is quite popular, and is also a scripting language. I think it's farther away, syntax wise, than Python, but I've never really used it so I can't promise that. I know it has at least GTK bindings.
Perl has a lot of resources and bindings, but isn't as easy to read as PHP since you have to learn the special variables like #_. It was never really my cup of tea.
You do have the C/C++ stuff, and Java has it's large library. You may want to go that way since you say you've used C++ before. If you're on a Mac (or willing to use GNUStep) you could go with Objective-C/Cocoa. That's getting rather far from PHP syntax though.
All of these languages have database connections. You don't mention what platform you're on.
But for easy to work with, quick to pick up, works all sorts of places, and can definitely do GUIs... Python would be a good choice to look at.
You could always use PHP. It is a decent command-line / programing application. Other than that all I can say is that you knowledge of database access and storage will be helpful but by the end of the day you are going to need to learn a new language.
Most languages have libraries for database access. Just pick one that you like the feel of. It is also a good idea to choose one that is popular (for the community support) and free libraries are always nice. Also look for good documentation and one that is fairly standard.
A nice thing to know is that javascript and php syntax are very similar to many other languages. (Javascript looks almost identical to C and C++). Just read the main language tutorial then the database API tutorial and you should be good to go
I have code that is straight forward, but intensive. I would like to compile it into a PHP extension. What is the simplest / best practice for going about that?
I have a Java background, so writing the C / C++ code shouldn't be an issue. I would, however, like to avoid any pitfalls along the way.
A simple tutorial or walk through is something I haven't been able to find. Plenty of examples, but nothing aimed at someone new to doing this.
It looks like a more arduous road then I though, especially with all the lovely C quirks.
The PHP Manual and this Zend article seem to cover quite a bit of ground in regards to writing PHP extensions.
The the tag info for php-extension and in particular, this answer, also linked from there.
You should try SWIG for generating wrapper for your source code, and then create your extension. It supports various languages like Java, C/C++ and makes it as simple as possible.
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 8 years ago.
Improve this question
I'm planning on moving to Python and I have a couple of additional questions along with the title:
did you have more fun with python?
are you as productive as when you're using PHP?
what made you change to python?
Would you do a project again in PHP? If so, why?
Your answers would really be useful for us PHP devs wanting something more I guess :)
Thanks in advance!
I was a PHP dev for about 5 years before switching to Python almost exclusively a year ago. The experience has been a mostly positive one; I'll answer your questions but also list a few gotchas I ran into.
Definitely. I continually find surprisingly powerful features/expressions in Python that do a great deal in a small amount of code (yet still being more readable than Perl).
Far more productive. It might just be my style, but Python's functional programming tools, generator expressions, list comprehensions, etc. allow me to accomplish tasks correctly with less code and less time invested than PHP.
I had an analytics project that needed a powerful stats package, so I went with Python+numpy. Then I found Turbogears and loved the syntax. Eventually I discovered coroutines and cooperative multitasking, and there's no going back. I use bottle, gevent, and gunicorn to crank out lean, fast, scalable web apps in record time.
Not if I could help it. PHP's verbose "everything is a long-named function call" syntax is just hard on my eyes at this point. I find it tedious to optimize as well (every page load reinterprets the source code in a default configuration).
Here are a few of the gotchas to be aware of:
For cheap, low-traffic sites, it's much harder to find a web host with a good python environment.
Apache isn't really a typical setup for Python in my experience. Python webapps are usually daemons that are exposed to the public with a reverse proxy webserver in front (nginx is very common). A number of corporate environments balk at new-fangled technology like nginx. It also takes some adjustment to think about your webapps as daemons, and it can take some effort at first to get your daemonizing correct and consistent.
If you use mysql, you will have some pain switching for a while. There just isn't a Python mysql library that is highly compatible with PHP-style mysql queries. For example, most of them don't use the simple "?" syntax for parameterized queries, so you can't just paste your queries over (you have to use printf-style "%s", etc.). Also, just the fact that you actually have to choose and install a mysql library is an extra step over PHP. This no longer bothers me, since I don't use mysql anymore anyway.
This is a broad topic with much, much more to say, but I hope this was helpful.
I'll try my best to answer your questions as best I can:
Did you have more fun with python?
I really enjoy how minimalist python is, having modules with non-redundant naming conventions is really nice. I found this to be especially convenient when reading/debugging other peoples code.
I also love all of the python tricks to do some very elegant things in a single line of code such as list comprehensions and the itertools library.
I tend to develop my applications using mod_wsgi and it took some time to wrap my head around writing thread-safe web applications, but it was really worth it.
I also find unicode to be much less frustrating with python especially with python 3k.
are you as productive as when you're using PHP?
For simple websites python can be less fun to setup and use. One nice feature of PHP that I miss with python is mixing PHP and HTML in the same file. Python has a lot of nice template languages that make this easy as well, but they have to be installed.
what made you change to python?
I became frustrated with a lot of the little nuances of PHP such as strange integer and string conversions and so forth. I also started to feel that PHP was getting very bloated with a lot of methods with inconsistent naming schemes. I was referring to the PHP documentation quite frequently despite having a large portion of the php library memorized.
Would you do a project again in PHP? If so, why?
I would develop a PHP project again, it has a lot of nice features and a great community. Plus I have a lot of experience with PHP. I'd prefer to use python, but if the client wants PHP I'm not going to force something they don't want.
Well, I started with PHP, and have delved into Python recently. I wouldn't say that I've "moved to", but I do use both (still PHP more, but a fair bit of Python as well).
I wouldn't say that I have more "fun" with Python. There are a lot of really cool and easy things that I really wish I could take to PHP. So I guess it could be considered "fun". But I still enjoy PHP, so...
I'm more productive with PHP. I know PHP inside and out. I know most of the little nuances involved in writing effective PHP code. I don't know Python that well (I've maybe written 5k lines of Python)... I know enough to do what I need to, but not nearly as in-depth as PHP.
I wanted to try something new. I never liked Python, but then one day I decided to learn the basics, and that changed my views on it. Now I really like some parts (and can see how it influences what PHP I write)...
I am still doing PHP projects. It's my best language. And IMHO it's better than Python at some web tasks (like high traffic sites). PHP has a built in multi-threaded FastCGI listener. Python you need to find one (there are a bunch out there). But in my benchmarks, Python was never able to get anywhere near as as fast as PHP with FastCGI (The best Py performed it was 25% slower than PHP. The worst was several hundered times, depending on the FCGI library). But that's based on my experience (which admittedly isn't much). I know PHP, so I feel more comfortable committing a large site to it than I would PY...
I run a self-developed private social site for 100+ users. Python was absolutely fantastic for making and running this.
did you have more fun with python?
Most definitely.
are you as productive as when you're using PHP?
Mostly yes. Python coding style, at least for me is so much quicker and easier. But python does sometimes lack in included libraries and documentation over PHP. (But PHP seems second to none in that reguard). Also requires a tad more to get running under apache.
what made you change to python?
Easier to manage code, and quicker development (A good IDE helps there, I use WingIDE for python), as well as improving my python skills for when I switch to non-web based projects.
Would you do a project again in PHP? If so, why?
Perhaps if I were working on a large scale professional project. PHP is so ubiquitous on the web A company would have a much easier time finding a replacement PHP programmer.
Last year I switched job to get away from PHP and work in Python. I'm very much satisfied with the decision I made :)
To answer the individual questions:
did you have more fun with python?
Yes!
are you as productive as when you're using PHP?
More productive I'd say. But the overall increased experience in programming also had something to with that.
what made you change to python?
You are not expected to be a jack of all trades in non-PHP jobs. (Photoshop/Web Design/Flash is required for many PHP jobs, and I hate Flash). And I liked Python/Django a lot.
4. Would you do a project again in PHP? If so, why?
If it's small stuff that's better done without any framework, then yes.
I've never really worked with PHP (nothing major) and come from the .NET world. The project I am currently on requires a lot of Python work and I must say I love it. Very easy and "cool" language, ie. FUN!
.NET will always be my wife but Python is my mistress ;)
yes
yes
curiosity, search for better languages, etc. (actually, I learned them somewhat in parallel many years ago)
yes, if a project requires it explicitly
disclaimer: I never really moved from php.
did you have more fun with python?
Yes. Lot more.
are you as productive as when you're using PHP?
No. I think more.
what made you change to python?
Django.
Would you do a project again in PHP? If so, why?
Only if it is required.
I have a fairly large object-oriented php 5 project, and as part of a change impact analysis, I'd like to compile a report on the usage of each existing class throughout the project.
It would help me immensely if I could find an existing tool that will analyze all the files in my project and generate some sort of report that lists, for example, all the class names of objects instantiated for each class in the project, and allow me to at least search this easily and quickly.
Any help here would be appreciated!
Check out nWire for PHP. It analyzes your code and recognizes such associations. It is built as an interactive tool, not as a reporting tool, but, if you insist, you can still connect to its' database (it uses H2, which is SQL compatible) and use an external reporting tool.
IMO Zend has some profiling tools that do just that, Or you can extrapolate this information from their Accelerator log.
Or try this with XDEBUG
Xdebug can trace your code and create code coverage statistics. There are additional tools like Spike PHPCoverage, which can generate nicely formatted reports, but since these are intended for test-coverage, it'll just give you a boolean result (eg. line of code is used or not used). You probably want a more detailed view (eg. how many times is it used).
Another option is to use the function trace feature of Xdebug. This will give you a detailed report of the actual call graph. You can determine which files was used the most from this. You'll need to write a parser for the data manually, but that shouldn't be too hard.
Finally, you could do the same thing with a static call graph. There are some tools available for php. Here are a few:
http://www.doxygen.nl/
http://phpcallgraph.sourceforge.net/
http://www.bytekit.org/
Again, you probably need to do some additional manual parsing on the output from those tools, to get something that applies to your use case.
The clever guys at Particletree, the same people behind the functionally and aesthetically gorgeous Wufoo often publish and release their PHP toolsets and utilities, the most recent of which being their PHP Quick Profiler. As you can probably tell, I have a huge amount of respect for those guys and love the stuff that they do.
A good PHP profiler is often hard come by, and PQP is most certainly the best I've come across. That said, nearly all of the various application frameworks have some form of profiling system, humble or otherwise, but none as nearly as in-depth and helpful as PQP. However, I usually find that the framework profiling tools are more linked into the code automatically, and if you use the framework's standard libraries then you'll have to do a lot less implementation with the profiling tool (this is definitely the case with CodeIgniter). But if you want that extra bit of power and flexiblity, PQP is great.
Let me know if you find anything better - I'd love to see it!
Jamie
I have looked at a good deal of other peoples source code and other open source PHP software, but it seems to me that almost nobody actually uses PEAR.
How common is PEAR usage out in real world usage?
I was thinking that maybe the current feeling on frameworks may be affecting its popularity.
PHP programmer culture seems to have a rampant infestation of "Not Invented Here" syndrome, where everyone appears to want to reinvent the wheel themselves.
Not to say this applies to all PHP Programmers, but them doing this apparently far too normal.
Much of the time I believe its due to lack of education, and that combined with difficulty of hosting providers providing decent PHP services.
This makes getting a workable PEAR installation so much more difficult, and its worsened by PHP's design structure not being favorable to a modular design.
( This may improve with the addition of namespaces, but have yet to see ).
The vast majority of PHP code I see in the wild is still classic amateur code interpolated with HTML, and the majority of cheap hosting that PHP users inevitably sign up for doesn't give you shell access.
In my (limited) experience, every PEAR project that was potentially interesting had major points against it:
Code is targetted at the widest audience possible. There are hacks in place all over the place to deal with old/unsupported PHP versions. New useful features are ignored if they can't be emulated on older versions, meaning you end up lagging behind the core language development.
Any given project tends to grow until it solves everyone's problem with a single simple include. When your PHP interpreter has to process all of that source code on every page hit (because the authors may not have designed it to be opcode-cache-friendly), there is a measurable overhead for processing thousands of unused lines of code.
Style was always inconsistent. It never felt like I was learning generalizable APIs like in other languages.
I used to use PEAR::DB at work. We discovered that most of our scripts spent their time inside PEAR code instead of our own code. Replacing that with a very simple wrapper around pgsql_* functions significantly reduced execution time and increased runtime safety, due to the use of real prepared statements. PEAR::DB used its own (incorrect at the time) prepared-statement logic for Postgres because the native pgsql_ functions were too new to be used everywhere.
Overall, I feel like PEAR is good as a "starter library" in many cases. It is likely to be higher quality code than any individual will produce in a short amount of time. But I would certainly not use it in a popular public-facing website (at least, not without a lot of tweaking by hand... maintaining my own fork).
Im my opinion PEAR is a good project but lacks people who want to work and keep working on it, most of the packages have inconsistent coding practices (I do not mean coding style) and there are lots of TODO's in the whole thing.
I find it useful sometimes for coding stuff I didn't know existed yet, like custom country validation functions and so on, otherwise I'm better served with any available framework out there (like CodeIgnite or Zend Framework).
The Pear library is the kinda stuff that just sits there, plugging away, with very little glory. If you are looking for something that it can do, and there's nothing more specifically targeted in the framework that you are using - go use it.
I've been working on a dating site for the last two years - and there's at least 65 pear-sourced files I've used, and are still live there today. Some, like the pager or html_Quickform will be overtaken by new code as it's updated, but for others there's just no need.
PEAR is not common, nor popular.
I tried to use PEAR so many times, but it lacks the umpphh to commit.
I prefer Zend Framework which takes the approach of 'loose' type, use only what you want.
PEAR is not common, nor popular — but it is good, and I'd recommend it to anyone.
(I do agree with Tom in that it doesn't feel like a single, unified, API; but then, this is PHP … one wouldn't like to see it getting above its station as an interpreted hack language now would one?!)