QB64.org Forum
Active Forums => QB64 Discussion => Topic started by: Pete on October 01, 2021, 12:33:41 pm
-
Organization is my middle, no first, no wait, I can't remember what part of my name it is. Well, that's about the truth. I started a little cat herding project to round up all the various aps, utilities, algorithms, etc., I've made in QB over the years, and what I'm finding is hundreds of files with hard to apps, many with hard to deduce old MS-DOS 8.3 names. It's reminding me of the Indiana Jones Arc of the Covenant lost in the government warehouse scene, only in my case, it's more of the "wherehouse." I mean I did start making build folders, about 8 years ago, but before that, I didn't even sort out builds.
So how about you guys? Anyone develop a system of tagging files or other methods so you have a good idea of what you programmed in your past?
Pete
-
I'm a "put it on GitHub"-organized kind of guy.
-
Well good for you, but no bueno for me. I tried putting my stuff on GitHub... and they told me to GitLost.
Pete
-
I used to use GitHub more than I do now. Right now, every bit of source is in my QB64 directory. Everything. Going several years back
-
Oh no Indy, you're going down my path! Watch out for the giant rock rolling up behind you! By the time you hit my age, you won't know where to find you car keys, let alone anything you programmed with your laptop keys.
When I was younger, I had a vivid memory of years of coding projects. When that memory ability began to fades with age, I got this odd feeling where I knew I had something, but I couldn't readily remember what it was called, or after many machine upgrades, what hard drive it might be on. That's why years ago, I just copied every file with a .bas extension to one BASIC folder, much like you put everything in a QB64 folder. So while I do know where all my work is, I now want to get those code files more indexed and accessible.
This thread brings back memories of a non-computer project I made in the 7th grade. I had this great idea of an all-purpose binder, that would organize all my school projects. I spent a couple of weeks designing and constructing it, and it was pretty impressive. The problem was, I only used it, as designed, a couple of times. I went right back to just stuffing papers everywhere, shortly after that! I really need to modify my Karma when it comes to this. In other words learn to love using what I create and much or more than creating it.
Pete
-
Unfortunately, my organization is a lot like Pete’s. I guess we’re both just Old Skool(tm) like that. I’ve got files and libraries and projects scattered across 20 hard drives, a dozen USB sticks, and more CDs and DVDs than I could ever use for coasters or frisbees!!
And, like Pete, I once worked on an organizational project to track all my stuff. I wrote a nice little database manager, had it track 50 or so various tags (sound, video, graphic, font, utility, math, game, database, ect…), and then I could use it to track my stuff. If I wanted a CircleFill routine, I’d click the tags for graphics, QB45 compatible, and utility and PRESTO — a link to the much smaller subset of files that met that criteria.
Designing the database, setting it up, making it interactive and pretty… all that was grand and amazing!! An excellent use of several weeks worth of time!!
Sorting through my legion of stuff, tagging it, and adding it to the database…. Not so much so. I enjoyed making the organizer, but when push came to shove, I just wasn’t enthused enough to use it afterwards…. I’m more of a program designer than I am a data entry type guy, by far.
So now even my grand SOS (Steve’s Organization System) sits in some folder, on some drive, or some archive on some CD or DVD, stored someplace in some room in my house…
-
There is a metaphysical theory we create what we think about. So 30 some odd years ago if something happened my wife didn't like, I used to tell her I didn't cause it, it was my evil twin, Steve! Also funny while Steve was posting here, I was making my post about the possibility of adding tagging to the IDE.
Anyway, evil part aside, just used for the bit, I have seen a lot of similarities between the two of us and I'm really glad Steve was able to start making libraries long before I had the time to pursue that system. Making libraries is a real pay it forward time saver. Still, and maybe some of you have had this same or similar experience, I have often passed on using a previously made routine and rather just coded, from scratch the same routine into a sub in the new project. I guess like Steve, I just like coding that much, too.
I kid Steve a lot, but I also have a lot of respect for him, his accomplishments, and contributions. I did not post this. It was my evil twin, Steve.
Pete
-
Hi
organization of files of code... in the years....
In the time of QBasic I have used a set of directory each for a project, while in the phase of developing I used to use a specific folder... Incorso (Ongoing) ....
BUT it doesn't save you from HDD crashes! So I used to make a copy on floppy, CD, external HDD (parallel port/ USB) following the evolution of technology of storing data.
Nowadays I find ancient to use a storage but still useful. The raising issue is the spreading (and duplication) of files in different paths (on the same storage or in different storage media) is a nightmare!
What file is the last version of that snippet? What features has this application ? Is that the main project that I have planned or is it a variant coming out from other thinkings about it?
For now because my coding is very few, cause of lack of time ( I have so many project ongoing, because some arguments showded in the forum stimulate my curiosity and my imagination), I haven't used a cloud system, that let you surely access to files anywhere. (For lost data issue you can have a backup on another cloud) So I'm using 2 folders out of QB64 folder. While in the QB64 folder I do anything, (I often save there code taken from QB64forum and from Discord) in the folder named MyQB64 I save my projects (there are folders for each of my projects plus Ongoing folder) and in the folder named QB64Forum (before it was QB64Store) I save snippet and programs posted from all you in the forum. For each of the more active contributors I have made a folder with subfolder for his programs.
I think that if I continue to write in QB64 I'll start to build library of code to use at need.
@Pete so you have all your codes in one folder! The idea of Steve is great. You can build a database program with index as Steve posted some days ago... Main indexes :namefile, dateodfcreation, size, kind of program, allyouthinkusefulfordistinguishitfromothercode.
GoodWeekend
@Pete
There is a metaphysical theory we create what we think about.
it is a branch of creationism
https://en.wikipedia.org/wiki/Creationism (https://en.wikipedia.org/wiki/Creationism)
-
[ This attachment cannot be displayed inline in 'Print Page' view ]
I use the last item listed in this .TXT file to get a listing of all Files and Folders of my "QB64 work" Folder (it makes the .TXT file).
When I can't remember where something is, I search the .TXT file with notepad++
I do the same with my other Basic folders.
-
I use multiple decimals within the file name to help search for the many files.
I started this a little ago when I found I could open files containing data but not remember the .exe which was responsible for creating the data in the first place. The decimals/periods in the name now give me the type of data and the program that created it.
So for example, if main program with all the menu's and calculation formulas is called "LightningStrikes.bas" and the data being stored is the number of occurrence by Ground Lightning, Cloud to Cloud, Forked, Injuries, Property Damage. The data gets stored in a File "Ground.Cloud.Forked.BI.PD.LightningStrikes.bas" .
I guess its like multiple tags and I'm not sure how many decimals/periods would eventually cause me head aches but so far I found a good description for the .exe and using decimals/periods in the .bas has helped in searching out what goes where.
-
I use multiple decimals within the file name to help search for the many files.
I started this a little ago when I found I could open files containing data but not remember the .exe which was responsible for creating the data in the first place. The decimals/periods in the name now give me the type of data and the program that created it.
So for example, if main program with all the menu's and calculation formulas is called "LightningStrikes.bas" and the data being stored is the number of occurrence by Ground Lightning, Cloud to Cloud, Forked, Injuries, Property Damage. The data gets stored in a File "Ground.Cloud.Forked.BI.PD.LightningStrikes.bas" .
I guess its like multiple tags and I'm not sure how many decimals/periods would eventually cause me head aches but so far I found a good description for the .exe and using decimals/periods in the .bas has helped in searching out what goes where.
Seems to me spaces would work just as well eg, "Ground Cloud Forked BI PD Lightning Strikes.bas" .
-
Major working projects and their associated files get a subdirectory. Everything else goes in a pile. One pile to rule them all, one pile to find them...etc., etc.
-
I've got a massive pile of programs floating on many pc's and disks ... Stuff that is useful gets copied to 'the new' pc and so lives on ...
Also I've found programs from which I really don't have an idea what they do.... specific tools probably. And also I suspect to have written programs twice to solve the same problem :-)
But I'm a bit more organized now, With a \BAS folder with all subfolder per program or group of programs.
And a tree of the same folders in a Projects share on a nas.
Stuff is backed up by Duplicati (a kind of Time Machine), which has really helped me out when I need to look back at an 'in-between version' and forgot to increase the version number and go on in a new file.
-
I used to have the same issue of source code all over many systems, OS and devices.
Since I got my ProBox drive bays and extra hard drives, I spent more than a month going through my source and placing them in their proper directories. Since I code in more than a dozen languages, I set my system up where I have a SourceCode directory, then by language (Assembler, BASH, C, C++, BASIC, Fortan, COBOL, COBOLScript, HTML, PHP, Python, PASCAL, JAVA<, JAVAScript, etc.) and within each folder, if I have different dialects of a language (like BASIC, I have QB64, TechBASIC, SmartBASIC, FreeBASIC, etc) and within each I have my source code by Projects.
For programs that are loose in my BASIC folder, I use the extension to tell me what they are: programa.qb64 or programa.tech or programa.sb, etc. And I have Geany set up so that the correct compiler is used.
If I use the QB64 IDE, then I can pull in any program with .qb64 extension and it still works.
All my includes for BASIC have the .INC extension, just for convenience. I do somethimes use .BI and .BM when warranted.
When I am developing something, I have other directories, such as Development, Test, Debug, and Release. I use this to freeze stable code before I screw it up with changes.
I hope that helps, but with 10,000+ source programs, I needed a better way to manage them. I do still have issues with duplicate programs, but I work on fixing that whenever I get a chance, unless the source is on my ARCHIVE and BACKUP hard drives. (If you do not have many hard drives, then you can use a folder labelled ARCHIVE and BACKUP).
It may sound like overkill, but this way, if I need code from TechBASIC to convert to QB64 or even in C, I know that I most likely only have one source to use. I wrote a BASH script that does check to see if I have duplicates of that source code, and resolve it at that time.
-
Nice topic by the way :-)
Maybe we should share the tools as well that we are using to conquer the mess we've created ...
Like I use Duplicati for backup and I would be helpless without Everything Search.
-
I've just 5 folders inside the qb64 installation folder
QB64Company -- Programs related/used at my working place
QB64GuiTools -- Lastest working release of my GuiTools
QB64GuiToolsDEV -- The current development branch of my GuiTools
QB64Library -- My Libraries Collection
QB64RhoSigma -- My private stuff, snippets, experiments ...
Each of this 5 folders has a maximum of 2 levels deep sub-folders to organize stuff, if it turns out it would be necessary to add another (3rd) level of sub-folders for a specific file/project, then I rather rethink if it wouldn'd eventually fit into an existing folder when looking at it from a different point of view, which worked out very well until now.
The 5 main folders are also in a private repository at GitHub, which is always up to date with the local content.
This way I've always a up to date codebase when doing stuff at work or at a friends computer. I just need to download and unpack QB64 and then download my stuff from GitHub into the QB64 folder and I'm ready to go.
Also I do pack the 5 main folders into a 7-Zip archive once a month and save it on 2 different USB flash sticks, because things saved in one place only are absolutly NOT SAVE.
-
I am compulsively organized, to a fault. When I play video games, I make spreadsheets and notes and track everything by hand. Right now I'm playing through Final Fantasy X for the nth time, and I have a spreadsheet showing instances remaining on each character's overdrive mode learning. For each type. So every time Lulu evades, I reduce the number by one in her "Dancer" row. Every time Wakka lands a status effect, same thing for his "Tactician" row. Sometimes when people watch me play games, it drives them crazy. Makes for very thorough speedrun routing, though!
I'm sure some here by now are familiar with (and annoyed by) my use of Notepad to code my big game project, but it's what I have to use, since I often do coding during downtime at work, and I can't use outside software. So I'm up around 6000 lines of code now, in Notepad. You don't manage that kind of thing without strict organization. All my routines are indexed and categorized, whitespace is used for visual clarity everywhere (the QB64 IDE would actually ruin this), and my folders are all where they are very much on purpose.
Being so organized takes time up front, but I reap pretty huge benefits later on. When I have to debug, it takes me maybe an hour or two at most, since I've already built in quite a bit of tools to let me see how data is being handled - such as a good old fashioned "debug menu" where I can scroll through the spawned entities and watch their data, edit player health, ammo, available weapons, turn on hitbox display, all during runtime. I even have a rudimentary level editor, minus some abstraction - it's really only meant for me to use.
-
I am compulsively organized, to a fault. When I play video games, I make spreadsheets and notes and track everything by hand. Right now I'm playing through Final Fantasy X for the nth time, and I have a spreadsheet showing instances remaining on each character's overdrive mode learning. For each type. So every time Lulu evades, I reduce the number by one in her "Dancer" row. Every time Wakka lands a status effect, same thing for his "Tactician" row. Sometimes when people watch me play games, it drives them crazy. Makes for very thorough speedrun routing, though!
I'm sure some here by now are familiar with (and annoyed by) my use of Notepad to code my big game project, but it's what I have to use, since I often do coding during downtime at work, and I can't use outside software. So I'm up around 6000 lines of code now, in Notepad. You don't manage that kind of thing without strict organization. All my routines are indexed and categorized, whitespace is used for visual clarity everywhere (the QB64 IDE would actually ruin this), and my folders are all where they are very much on purpose.
Being so organized takes time up front, but I reap pretty huge benefits later on. When I have to debug, it takes me maybe an hour or two at most, since I've already built in quite a bit of tools to let me see how data is being handled - such as a good old fashioned "debug menu" where I can scroll through the spawned entities and watch their data, edit player health, ammo, available weapons, turn on hitbox display, all during runtime. I even have a rudimentary level editor, minus some abstraction - it's really only meant for me to use.
@johannhowitzer I'm sincerely glad that method works for you, but if I took your approach, I have zero programs and a nice little 10 x 6 padded cell to reflect on my lack of productivity. So for me to have 6,000, 20,000, even 90,000 line projects means I have to give up all the back end time savings and run and gun my way through the prolect. When I was younger, I'd code in all caps, use single or two-letter variable names, no white space, and no comments. I miss those days. I'm more conventional now, but I find myself wishing I had more synergism between coding and organizing. I love to code, but hate to include comments. I feel like I would spend too much time commenting, indexing, and trying to figure out what I might need to revisit the project. Doing that totally kills my interest in creating the project. Fortunately, on the back end, I have little resistance to debugging or code modifying but it does drive me crazy to try to recall what the various parts of the project accomplish, or if it is even safe to run without changing files on my system. So my hat's off to you for being able to embrace the best of both worlds. From what I'm reading here, that's more the exception than the rule.
Pete