Question posted in 
looking for an “online ide open source” preferably working top of node.js
i want to store all javascript, html, css, sql into a database vs files.
it would be nice to see a version control, bug tracker, minified/no minified built into it.
i would be nice to see function / class / variable / generic comment box that is tied directly to current line in code. (more of a spreedsheet feel to it, code is in A column, and comments in B column).
it would be nice to be able to “order” code. as in this function, class, varibles, etc… get loaded first, and then this, and then this, etc.. by simply entering 1,2,3,4,5,6,etc… (think of it as C column in a spreedsheet)
it would be nice to have a “tree base” if you have ever dealt with “custom javascript menus” then (column C,D in spreedsheet), would tie into ordering of code.
it would be nice to have “styling rules” that can be ran across the code. directly from within the ide.
===============
long story
from the old oscommerce days of “hand coding” and using winmerge in attempt to see differences and apply updates to mods / addons. to the more newer zencart / opencart / joomla / phpbb were there is more of an easy auto install directly built into back end with update abilty. to github and constant reliance of version control software, but there are still some “hard coding” of files for certain things that does happen on individual sites. it would be nice, if i could click, “search for updates”, and then look at version control area of IDE. and see if i need to “hand code” around certain modifications i have made to core files heck even mods / addons. vs the standard code base. and then apply auto update to rest of code. without wiping out 20 other programs and a descent amount of other commands to get the job done.
who hates the never ending 1000 to 10,000 plus long lines of code per file and trying to search through to find some function or like? who hates trying to flip through 20 to 200 different plus files with each ranging form 5 lines of code to say 600 lines of code. that you got from notepadd++ -> CTRL + F -> search in folders/files. ((opening up a file that is 5 to 10 lines of code gets me))
who has issues of “naming convention” for files? from a prefix, to camelcase, to in certain directories / folders, to certain folder permissions, to specific type of file, what order and how you require / include another file into current file. all the large / huge software packages out there are different. it would be nice to just code up some things and type into a “title” / “description”, textinput / textarea in an IDE, and database all the code. most of the javascript, css, even html, is getting code crunched per say. and some were in the core files it is all being pushed into a compiled bigger single files per say during run time or during “cache” creation perhaps. what is big deal having all the code in database? when many of the files / folders may only have a couple lines of code in them, and about 20 lines of “copy right info”, and perhaps only 1 or 2 files in them?
if i can log into ide, i get my username or rather my “code signature” errr forum signature, email signature, etc…. attached, and i can adjust licensee type as needed. and have it database right along with the code. heck even per line, or function, vs blobs and blobs of lines of code.
if i make a bug comment / tracker issue / general comment on a piece of code. i could click a little tab or button…((see spreedsheet comment at top)) and this would be like E / F / G columns or some such. heck could have little mini “forum thread” or rather mini forum going for it. it would be right there in IDE. and IDE built directly into the larger project. and i would be able to see who dealt with code, perhaps who tested code, who approved code, when it was push/pulled, etc… there is to much “splitting” and folks dedicated to certain tasks, as a result things being split across multi programs / addons / mods / etc…
there seems to be always a “large debate” that kinda never ends… that i remember seeing for most all larger open source projects out there. when it comes to were to load javascript in the html file, and how much to load at very beginning vs dropping a large chunk perhaps at the very end of the html file and waiting for document.load to complete before loading javascript. if i can go down the code that is in the database. and order it and place it into “tree” style format. i might be able to see certain code getting nailed on heavy and certain parts of it needing to loaded at the very beginning through analyzing the page in browsers and like. and simply bam, bam, done… vs going back and adjust how each individual module / mod / addon / core files get read and processed, whitch most likely means copy/paste and adjust a good amount of code in multi files.
there always seems to be some sort of “third party” frame work / addin / mods, say jquery, or yui, or certain third party javascript files. many times i see initially full files and then a minified files, setting beside each other in some folder of the project. then eventually i start seeing these files getting torn apart and mini amount of code is left, in attempt to reduce overall javascript that is needed. normally a good amount gets customized javascript via “sniplets” tossed into a varibles vs actually using dedicated full framework. heck many times i see same like frameworks, but done by some other third party. they do same thing, but folks don’t know it. so they just toss the entire framework of what they know in. raises hand guilty as charged.
it would be nice, to have a built in ide, that auto scans functions, classes, variables, so when you type them in somewhere, “auto complete” might finish it out, to “tool tip” poping up, to being able to add “API reference” information. directly through IDE were the code is. vs going through another program. ya honestly… never much liked this type of info in various IDE’s over the years. more so when initially starting out in a bigger project that is new to me. but information does come in handy from time to time. notepad++, CTRL + F, find in folders…. only goes so far. in searching see how things were used. trying to dig info up on some html page that might be buried and hard to find… vs in the IDE.
i am looking at SAAS (software as a service) type of thing using HTML5 sql database built into browser and like abilities. so i can push the IDE, plus larger project itself, directly to the browser. and then go offline. and be able to shut down computer and at later time bring computer backup without internet and code, check for bugs, etc… it would be nice to login in one computer do some things, “do a save online but do not submit” option, were i can then log into another computer and go at it (assumption both computers were/are connected to internet).
it would be nice, to use something like “games” have it and/or some “video chats”, and doing a local lan type gaming. but instead. sharing code/updates/software itself that is stored in the browser sql database. longer shoot, windows 10 and how if you have more than 1 computer, if both computers are online and one of them has updates. the one that does not have updates, will pull from the one that does. so as to use less bandwidth of the internet. ((torrent like abilty per say)) if i can do a google docs multi person at same time working on spreed sheet. why not an IDE. there be multi chat/ paint like online programs that can do this.
i would think there would be an option on development site err IDE i mean, to click “minified javascript” or non minified. so as to make things easier to debug in browser. perhaps even selecting “what code gets minified and what does not… via “ordering” of code, and check mark for minified or not. with simply “couple new lines and comments” to make it easier to scroll right to given code in browser.
=============
i don’t mind starting from scratch. just hoping someone might know of something, that i simply have not came across. most of my searches for files vs database… is files… files… files… database big fat NO. but if i only need to run sql quires when ever a “code change happens” to build a handful of files. the overall load should be min. and hopefully faster, vs checking the cache constantly and rebuilding the cache files.
this is the complicated way… of getting more near “compiled” code with javascript, css, html/templates. at least that is my hopes for a faster running site, and less bandwidth used overall, hopefully reduced mess of programs needed to actually work on larger project. it will not be for all. but hopefully tie things all together. and again this is only part of the overall backend of the larger project. need to start some were this looks like best place to start, were all the code begins and ends.
2
Answers
rubs eyes partial answers... from same person...
""js-git"" looks to be a nice framework, with already addon's for going to memory, db, or html5 offline doing. with kickstarter going for it.
""tedit"" on github with google / firefox store apps.
demo on youtube showing js-git with a browser base front end. showing more offline doing, but i want to say it showed some additional info... history, tree, and like. i got lost / side tracked and unable to find code for the demo shown.
=============== WYSIWYG (what you see is what you get) editor / html editor. seems to be next issue.
i was hoping ""libra / online"" would be something for word/document app, but meh pc went to a standstill. unlike tinymce / ckeditor and like out there. and what ever stack is using.
time to find something half way descent for node.js
ECA (event condition action) aka triggers, if i can find something along those lines. that works worth more of a css buttons, and min javascript, and a little bit of html.
I read all of your question, and I think I have a reasonable understanding of how your workflow is, and how you want it to be.
It seems to me, that you can solve most of your issues with using a unix based operating system.
What you need seems to be: A styleguide, Grunt, Git, Unix
1: A styleguide will fix any problems you have with naming conventions
2: Grunt can fix browsersync between users and devices
3: Git fixes version control
4: unix* fixes the “too many files” problem
*The grep command is a god-send when it comes to searching for content inside a file. “Grep -Rl find_a_function /my_project_folder/” will give you every file with the text “find_a_function” in your project folder, and sub-folders. This negates most of the problems with naming convenstions, even though the styleguide should fix this. You can then use “grep -R find_a_code_part /my_project_folder/listed_folder/listed_folder/” to view the lines surrounding the code you wanted to find, to easily find the code you are looking for.
You can make every part of the page run locally, to avoid the “I can’t work without Internet” issues. In addition to this, you can use any IDE. Netbeans, Eclipse, PHP Storm or not an IDE at all. I found using vim is often faster than using an IDE, because there are shortcuts for almost anything you would want. There is also the “I make my own aliases in bash” aspect. So you can have short commands to be aliases for actual commands.
and all kinds of crazy shortcuts to make your life of developing easier.