I think they should just completely replace the command prompt with powershell. Maybe have a compatability mode that's ran by default if something launches a batch file using an 'old' file extension.
They don't support the 16 bit apis in 64 bit windows, right? I really hope they drop 32 bit with windows 10.
Sure they do. 16 bit code isn't supported, but that's a processor thing. The APIs haven't actually been 16-bit for a while, that's just when they were added, but some are still valid to call in 32-bit mode. In win7 32-bit I could run old 16-bit windows games and they worked. Games from the early 90s
I actually like PDFs. As long as they aren't forms you can't actually fill out, or 600 page, tends of megabyte monstrosities.
I cannot figure out how to byte64 encode the contents of a PDF and send the string through a JSON Response in Python. I can do it with every other file format under the sun but I cannot do it with PDF.
I actually like PDFs. As long as they aren't forms you can't actually fill out, or 600 page, tends of megabyte monstrosities.
I cannot figure out how to byte64 encode the contents of a PDF and send the string through a JSON Response in Python. I can do it with every other file format under the sun but I cannot do it with PDF.
Same issue I had the other day with these SHA1 hashing algos, you're probably not reading the file right.
You need to do a binary ready of the file. Binary is binary whether it's JPG or PNG or PDF or TXT.
What language are you using for file reading? PHP or Python?
bowen on
not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
Going for my second round of interviews for TEALS volunteering today. Hopefully I get placed with a school that I like. I'm still kind of curious how this whole thing is going to work, but I'm also pretty excited.
I actually like PDFs. As long as they aren't forms you can't actually fill out, or 600 page, tends of megabyte monstrosities.
I cannot figure out how to byte64 encode the contents of a PDF and send the string through a JSON Response in Python. I can do it with every other file format under the sun but I cannot do it with PDF.
Same issue I had the other day with these SHA1 hashing algos, you're probably not reading the file right.
You need to do a binary ready of the file. Binary is binary whether it's JPG or PNG or PDF or TXT.
What language are you using for file reading? PHP or Python?
Reading it in python. I do read the file as bytes and then base64encode it but it just wouldn't work properly for me for some reason.
I actually like PDFs. As long as they aren't forms you can't actually fill out, or 600 page, tends of megabyte monstrosities.
I cannot figure out how to byte64 encode the contents of a PDF and send the string through a JSON Response in Python. I can do it with every other file format under the sun but I cannot do it with PDF.
Same issue I had the other day with these SHA1 hashing algos, you're probably not reading the file right.
You need to do a binary ready of the file. Binary is binary whether it's JPG or PNG or PDF or TXT.
What language are you using for file reading? PHP or Python?
Reading it in python. I do read the file as bytes and then base64encode it but it just wouldn't work properly for me for some reason.
Odd.
Want to post your code? I don't see anything that would necessarily cause a gotcha.
not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
Started poking at Grails after getting familiar with the Groovy syntax, and I'm really impressed. I don't know shit about actual performance metrics, but Grails as a framework is really solid.
I actually like PDFs. As long as they aren't forms you can't actually fill out, or 600 page, tends of megabyte monstrosities.
I cannot figure out how to byte64 encode the contents of a PDF and send the string through a JSON Response in Python. I can do it with every other file format under the sun but I cannot do it with PDF.
Same issue I had the other day with these SHA1 hashing algos, you're probably not reading the file right.
You need to do a binary ready of the file. Binary is binary whether it's JPG or PNG or PDF or TXT.
What language are you using for file reading? PHP or Python?
Reading it in python. I do read the file as bytes and then base64encode it but it just wouldn't work properly for me for some reason.
Odd.
Want to post your code? I don't see anything that would necessarily cause a gotcha.
I've kind of abandoned the code (this was a few weeks/months ago) as the PDF wasn't the priority at the time.
What library were you using for the PDF writing?
The one time I did anything PDF in Python I used ReportLab and it worked really well.
Granted, I was just throwing pure text in there.
But when I create the model and don't include a description it will default to an empty string every time, allowing it to get created even if you don't pass in the description. So, for example, this:
For both string-based and non-string-based fields, you will also need to set blank=True if you wish to permit empty values in forms, as the null parameter only affects database storage (see blank).
Why would properties of objects that I initialize in a collection suddenly lose their values as soon as I pull the individual object out of the collection?
Somehow, this doesn't work. When I pull the individual objects out, and try to reference the properties I initialized, it says the properties have never been initialized. For example, later in the program, getting an individual SP.ListItem from this collection and referencing item.File.Name throws a server exception.
In order to fix this problem, I have to do the following, which sucks:
CamlQuery query = new CamlQuery();
query.ViewXml = "<View></View>";
SP.ListItemCollection listItems = currentList.GetItems(query);
clientContext.Load(listItems,
items => items.Include(
thisItem => thisItem.Id));
clientContext.ExecuteQuery();
List<SP.ListItem> itemsToAdd = new List<SP.ListItem>();
// Add each ListItem I want to work with to this new List
// Use the first foreach loop to call Load on each of our items
// This is done to ensure only one ExecuteQuery is called
foreach(SP.ListItem item in itemsToAdd)
{
clientContext.Load(item,
thisItem => thisItem.File,
thisItem => thisItem.File.ServerRelativeUrl,
thisItem => thisItem[Properties.Resources.FileGuidInternalName],
thisItem => thisItem[Properties.Resources.FileNameInternalName],
thisItem => thisItem.ContentType,
thisItem => thisItem.ContentType.Name,
thisItem => thisItem.File.Name,
thisItem => thisItem.Id);
}
clientContext.ExecuteQuery();
// Actually work with my ListItem objects down here
What library were you using for the PDF writing?
The one time I did anything PDF in Python I used ReportLab and it worked really well.
Granted, I was just throwing pure text in there.
The product I was using was DynamicPDF. This was all in C#.
Sorry, I was actually aiming that honky, I just posted my reply way late.
Oh I'm sorry Tofy, I wasn't writing the PDFs in a sense. I was opening the PDF and I wanted to send the bytes to the front end client.
0
Options
GnomeTankWhat the what?Portland, OregonRegistered Userregular
Atom seems neat until you realize it runs like a dog, consumes way too much memory, and has trouble doing heavy multi-tasking loads. Not really shocking when you figure it's written in HTML and JavaScript, while Sublime is written in C++.
Argh, my gitignore is being mean to me. How do I ignore dot folders? I don't want IntelliJ's stuff in the repo. I have other projects where I successfully ignore folders with just "foo" in the .gitignore, but git won't ignore ".idea".
edit: duh, they weren't being ignored because the files were already tracked by git. Untracked them and now it works as intended.
Atom seems neat until you realize it runs like a dog, consumes way too much memory, and has trouble doing heavy multi-tasking loads. Not really shocking when you figure it's written in HTML and JavaScript, while Sublime is written in C++.
I've been running it all day and it's using 32MB of memory. When was the last time you used it?
Is there a reason you can't just say "go get node, go get npm, go get bower, use your text editor of choice" on windows / OSX / linux, and use railsinstaller for that? When it comes to deploy time, sure, the environment will be more important, but for just basic web stuff it seems like all you really need to make is some text files so you could start with notepad/vi/textedit -- there's railsinstaller for that stuff (though I don't know much about rails myself -- is it really picky about the exact way it's set up / has a bunch of other dependencies / that sort of thing?)
I've had mixed experiences with rails installer on windows. I haven't given it a shot in a while, so it may be just fine.
But I definitely want a proven-working environment to hand out to people, because I'd rather they figure out web development instead of infrastructure provisioning. Let's call it beyond the scope of the class.
Going for my second round of interviews for TEALS volunteering today. Hopefully I get placed with a school that I like. I'm still kind of curious how this whole thing is going to work, but I'm also pretty excited.
I didn't even know this was a thing. Now I want to look into it too. Thanks!
It's a little awkward having our old boss still be here even though he's been pretty much relieved of his management status. In his contract it says that he must remain for 90 days and get paid for the 90 days after the company lays him off or he quits.
Long story short he's trying to make the software we're currently designing be marketable and not JUST for our company. That wasn't in line with our company's goals and we're way behind schedule so they made some cuts. The problem is that he is still here and comes in to see what decisions we've made on certain apps and wanting to see progress and stuff and I keep getting the feeling he's taking mental notes on everything we're building and all of the gains/changes we've made to make development happen quicker.
It's a little awkward having our old boss still be here even though he's been pretty much relieved of his management status. In his contract it says that he must remain for 90 days and get paid for the 90 days after the company lays him off or he quits.
Long story short he's trying to make the software we're currently designing be marketable and not JUST for our company. That wasn't in line with our company's goals and we're way behind schedule so they made some cuts. The problem is that he is still here and comes in to see what decisions we've made on certain apps and wanting to see progress and stuff and I keep getting the feeling he's taking mental notes on everything we're building and all of the gains/changes we've made to make development happen quicker.
That means the company can sell the product. This is how a lot of big companies function. They make an internal system, then begin designing it for resale to make money.
Nothing odd about that. Having that design in mind when you're designing the software from the get go makes a hell of a lot of difference in doing it in the end. There's probably a market, your manager probably knows about the market, you'd be dumb if you didn't direct a product that way during development. You see this in companies like Google/3M/Xerox. A lot of the Google projects you use every day were internal at one point. Gmail and google drive I think were the big ones.
That is... so long as it didn't interfere to heavily with deadlines or code.
If you're creating hundreds of extra hours of work for yourself, that's one thing. But if you're designing, say, a table to store data other than what you've got there, that's a good thing.
But something tells me the design path he chose for this project was adding a lot of extra time and making you guys miss deadlines because of it.
not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
Atom seems neat until you realize it runs like a dog, consumes way too much memory, and has trouble doing heavy multi-tasking loads. Not really shocking when you figure it's written in HTML and JavaScript, while Sublime is written in C++.
I've been running it all day and it's using 32MB of memory. When was the last time you used it?
Very recently. And memory is the least of my worries with Atom...it's incredibly slow. Do a multi-file search across several gigs of files sometime with Atom, and then do it with Sublime, and watch how much faster, and smoother, Sublime does that. Who cares if the package manager in Atom is nicer? How often do you use the package manager? I use multi-file search every day, multiple times a day.
When I want big and bulky I have Visual Studio. I want ridiculously lean and fast from my text editor. Same reason I don't use Notepad++, because it's bloated and slow. For how much Atom tries to bite Sublime's style (and lets face it, the Atom team has basically just taken every idea that Sublime popularized, whole hog), the one thing it doesn't do that actually matters is be lean and compact.
It's a little awkward having our old boss still be here even though he's been pretty much relieved of his management status. In his contract it says that he must remain for 90 days and get paid for the 90 days after the company lays him off or he quits.
Long story short he's trying to make the software we're currently designing be marketable and not JUST for our company. That wasn't in line with our company's goals and we're way behind schedule so they made some cuts. The problem is that he is still here and comes in to see what decisions we've made on certain apps and wanting to see progress and stuff and I keep getting the feeling he's taking mental notes on everything we're building and all of the gains/changes we've made to make development happen quicker.
That means the company can sell the product. This is how a lot of big companies function. They make an internal system, then begin designing it for resale to make money.
Nothing odd about that. Having that design in mind when you're designing the software from the get go makes a hell of a lot of difference in doing it in the end. There's probably a market, your manager probably knows about the market, you'd be dumb if you didn't direct a product that way during development. You see this in companies like Google/3M/Xerox. A lot of the Google projects you use every day were internal at one point. Gmail and google drive I think were the big ones.
That is... so long as it didn't interfere to heavily with deadlines or code.
If you're creating hundreds of extra hours of work for yourself, that's one thing. But if you're designing, say, a table to store data other than what you've got there, that's a good thing.
But something tells me the design path he chose for this project was adding a lot of extra time and making you guys miss deadlines because of it.
Well the big thing is that he wanted us to finish the product in one way and that didn't line up with our company's wants/needs. He was going to take this product, split off from our company, and make it his own company/project.
Atom seems neat until you realize it runs like a dog, consumes way too much memory, and has trouble doing heavy multi-tasking loads. Not really shocking when you figure it's written in HTML and JavaScript, while Sublime is written in C++.
I've been running it all day and it's using 32MB of memory. When was the last time you used it?
Very recently. And memory is the least of my worries with Atom...it's incredibly slow. Do a multi-file search across several gigs of files sometime with Atom, and then do it with Sublime, and watch how much faster, and smoother, Sublime does that. Who cares if the package manager in Atom is nicer? How often do you use the package manager? I use multi-file search every day, multiple times a day.
When I want big and bulky I have Visual Studio. I want ridiculously lean and fast from my text editor. Same reason I don't use Notepad++, because it's bloated and slow. For how much Atom tries to bite Sublime's style (and lets face it, the Atom team has basically just taken every idea that Sublime popularized, whole hog), the one thing it doesn't do that actually matters is be lean and compact.
I'm in the package manager more than I am searching over several gigs worth of data. :P
+1
Options
mightyjongyoSour CrrmEast Bay, CaliforniaRegistered Userregular
It's amazing how much of a difference -Dpoop=1234 vs. -Dpoop="1234" makes when you're using them in a function call =/.
ive been in the rabbit hole for 2 months part time building a completely browser based JS excel model evaluator
and finally got my first proof working in an extremely large real world excel workbook.... hundreds of thousands of cells, paragraph formulas, all computing accurately in real time
it was way harder than I expected it to be
but runs surprisingly fast in chrome. i can evaluate the entire workbook in under a few seconds, which matches C# performance for an equivalent library
It was something like this, not exactly, but my php install definitely did not give me an error. It just happily churned along until I tried to access Messages.
not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
This is going to get drastically more complex, but after I get the first bit working, the rest should be conceptually the same and fall into place... I hope. So, here's what I'm going for initially.
Return the account docs, sorted by score. Score is the number of years of experience in a single specified subject. If they have no experience in that subject, that's fine, the score is 0 but they should still be in the results. Any other subjects they also happen to have experience in should be ignored.
I've tried the following (assuming I'm looking for experience in subject 1), but all scores are always 0, so I quite obviously have no fucking idea what I'm doing. It makes sense to me that this wouldn't work because there's nothing obviously telling it to iterate over the nested experiences, but I cannot find any documentation or examples which clearly do so (which concerns me... maybe you can't do what I want to do). I realize these do not specify the sort order.
Longer term there will be more nested fields on the account doc and the score will not be a raw value, but some light math will be done on some of the values. I am fine with solutions which use other ways of structuring the data. All I care about getting back in the results in the end is sorted by score with highest score first and account.id. Everything else being stored is purely for searching on, then I'm going to take top N id's and SELECT * FROM some_table WHERE id IN [<those ids>].
Posts
They don't support the 16 bit apis in 64 bit windows, right? I really hope they drop 32 bit with windows 10.
I cannot figure out how to byte64 encode the contents of a PDF and send the string through a JSON Response in Python. I can do it with every other file format under the sun but I cannot do it with PDF.
Same issue I had the other day with these SHA1 hashing algos, you're probably not reading the file right.
You need to do a binary ready of the file. Binary is binary whether it's JPG or PNG or PDF or TXT.
What language are you using for file reading? PHP or Python?
No issues that I'm aware of, I am connecting immediately.
Let me know if it persists, probably just the tubes.
Reading it in python. I do read the file as bytes and then base64encode it but it just wouldn't work properly for me for some reason.
Odd.
Want to post your code? I don't see anything that would necessarily cause a gotcha.
I've kind of abandoned the code (this was a few weeks/months ago) as the PDF wasn't the priority at the time.
The one time I did anything PDF in Python I used ReportLab and it worked really well.
Granted, I was just throwing pure text in there.
I have a model that has the following:
But when I create the model and don't include a description it will default to an empty string every time, allowing it to get created even if you don't pass in the description. So, for example, this:
Will print an empty string even though I have it set as null=False. Do I have to update the default keyword manually to say:
e: Hmm still works without adding the field.
Why would properties of objects that I initialize in a collection suddenly lose their values as soon as I pull the individual object out of the collection?
Microsoft confuses me.
Somehow, this doesn't work. When I pull the individual objects out, and try to reference the properties I initialized, it says the properties have never been initialized. For example, later in the program, getting an individual SP.ListItem from this collection and referencing item.File.Name throws a server exception.
CamlQuery query = new CamlQuery(); query.ViewXml = "<View></View>"; SP.ListItemCollection listItems = currentList.GetItems(query); clientContext.Load(listItems, items => items.Include( thisItem => thisItem.File, thisItem => thisItem.File.ServerRelativeUrl, thisItem => thisItem[Properties.Resources.FileGuidInternalName], thisItem => thisItem[Properties.Resources.FileNameInternalName], thisItem => thisItem.ContentType, thisItem => thisItem.ContentType.Name, thisItem => thisItem.File.Name, thisItem => thisItem.Id)); clientContext.ExecuteQuery();In order to fix this problem, I have to do the following, which sucks:
CamlQuery query = new CamlQuery(); query.ViewXml = "<View></View>"; SP.ListItemCollection listItems = currentList.GetItems(query); clientContext.Load(listItems, items => items.Include( thisItem => thisItem.Id)); clientContext.ExecuteQuery(); List<SP.ListItem> itemsToAdd = new List<SP.ListItem>(); // Add each ListItem I want to work with to this new List // Use the first foreach loop to call Load on each of our items // This is done to ensure only one ExecuteQuery is called foreach(SP.ListItem item in itemsToAdd) { clientContext.Load(item, thisItem => thisItem.File, thisItem => thisItem.File.ServerRelativeUrl, thisItem => thisItem[Properties.Resources.FileGuidInternalName], thisItem => thisItem[Properties.Resources.FileNameInternalName], thisItem => thisItem.ContentType, thisItem => thisItem.ContentType.Name, thisItem => thisItem.File.Name, thisItem => thisItem.Id); } clientContext.ExecuteQuery(); // Actually work with my ListItem objects down hereWhy is this arcane ritual necessary?
I like it. I really like it. Package manager is much easier than Sublime and it just overall feels smoother.
Sorry, I was actually aiming that honky, I just posted my reply way late.
edit: duh, they weren't being ignored because the files were already tracked by git. Untracked them and now it works as intended.
I've been running it all day and it's using 32MB of memory. When was the last time you used it?
But I definitely want a proven-working environment to hand out to people, because I'd rather they figure out web development instead of infrastructure provisioning. Let's call it beyond the scope of the class.
I didn't even know this was a thing. Now I want to look into it too. Thanks!
Long story short he's trying to make the software we're currently designing be marketable and not JUST for our company. That wasn't in line with our company's goals and we're way behind schedule so they made some cuts. The problem is that he is still here and comes in to see what decisions we've made on certain apps and wanting to see progress and stuff and I keep getting the feeling he's taking mental notes on everything we're building and all of the gains/changes we've made to make development happen quicker.
That means the company can sell the product. This is how a lot of big companies function. They make an internal system, then begin designing it for resale to make money.
Nothing odd about that. Having that design in mind when you're designing the software from the get go makes a hell of a lot of difference in doing it in the end. There's probably a market, your manager probably knows about the market, you'd be dumb if you didn't direct a product that way during development. You see this in companies like Google/3M/Xerox. A lot of the Google projects you use every day were internal at one point. Gmail and google drive I think were the big ones.
That is... so long as it didn't interfere to heavily with deadlines or code.
If you're creating hundreds of extra hours of work for yourself, that's one thing. But if you're designing, say, a table to store data other than what you've got there, that's a good thing.
But something tells me the design path he chose for this project was adding a lot of extra time and making you guys miss deadlines because of it.
Very recently. And memory is the least of my worries with Atom...it's incredibly slow. Do a multi-file search across several gigs of files sometime with Atom, and then do it with Sublime, and watch how much faster, and smoother, Sublime does that. Who cares if the package manager in Atom is nicer? How often do you use the package manager? I use multi-file search every day, multiple times a day.
When I want big and bulky I have Visual Studio. I want ridiculously lean and fast from my text editor. Same reason I don't use Notepad++, because it's bloated and slow. For how much Atom tries to bite Sublime's style (and lets face it, the Atom team has basically just taken every idea that Sublime popularized, whole hog), the one thing it doesn't do that actually matters is be lean and compact.
Well the big thing is that he wanted us to finish the product in one way and that didn't line up with our company's wants/needs. He was going to take this product, split off from our company, and make it his own company/project.
But that is still very frustrating. I'm using sublime over atom at the moment.
I'm in the package manager more than I am searching over several gigs worth of data. :P
sigh.
Apparently that's 100% valid php, but doesn't work.
jetbrain's IDEs are worth the investment.
and finally got my first proof working in an extremely large real world excel workbook.... hundreds of thousands of cells, paragraph formulas, all computing accurately in real time
it was way harder than I expected it to be
but runs surprisingly fast in chrome. i can evaluate the entire workbook in under a few seconds, which matches C# performance for an equivalent library
Been there, done that.
Which version of PHP are your working with?
This code here: 3v4l.org/900ig seems to work ok in PHP5+
Oo\ Ironsizide
<?php class Foo { public $bar = 0; private $bar2 = 0; public function makeItFive($variable) { $this->$variable = 5; return $this->$variable; } } $test = new Foo(); print $test->makeItFive("bar"); print $test->makeItFive("bar2");Both calls return the expected 5.
Oo\ Ironsizide
It works that way, yeah. But I had an internal class I was calling a constructor on and passing a UUID too.
It took the UUID and assigned it to the variable instead.
Let me see if I can reproduce it again.
It was something like this, not exactly, but my php install definitely did not give me an error. It just happily churned along until I tried to access Messages.
Any ElasticSearch experts here? I am in way the fuck over my head and have no idea what I'm doing.
I've got documents with the following mapping
{"account": { "properties": { "id": {"type": "integer"}, "email": {"type": "string"}, "experiences": { "type": "nested", "properties": { "years": {"type": "integer"}, "subject": {"type": "integer"}, } } } }}This is going to get drastically more complex, but after I get the first bit working, the rest should be conceptually the same and fall into place... I hope. So, here's what I'm going for initially.
Return the account docs, sorted by score. Score is the number of years of experience in a single specified subject. If they have no experience in that subject, that's fine, the score is 0 but they should still be in the results. Any other subjects they also happen to have experience in should be ignored.
I've tried the following (assuming I'm looking for experience in subject 1), but all scores are always 0, so I quite obviously have no fucking idea what I'm doing. It makes sense to me that this wouldn't work because there's nothing obviously telling it to iterate over the nested experiences, but I cannot find any documentation or examples which clearly do so (which concerns me... maybe you can't do what I want to do). I realize these do not specify the sort order.
{"query": { "function_score": { "script_score": { "script": "if(doc['experiences.subject'] == 1){_score = _score + doc['experiences.years'].value} else {0}" } } }} {"query": { "function_score": { "script_score": { "script": "if(_source.experiences.subject == 1){_score = _score + _source.experiences.years} else {0}" } } }}Longer term there will be more nested fields on the account doc and the score will not be a raw value, but some light math will be done on some of the values. I am fine with solutions which use other ways of structuring the data. All I care about getting back in the results in the end is sorted by score with highest score first and account.id. Everything else being stored is purely for searching on, then I'm going to take top N id's and SELECT * FROM some_table WHERE id IN [<those ids>].