sheet2graph command-line program

Transcript

we saw before that pandas has an interesting option here in

the read excel that says read from a local file stem or

url right so this would be cool if our users

could like maybe they have some like google drive

excel or like some kind of spreadsheet and they can just

put the url there instead of downloading it and then

graphing it especially they have a lot of them so

we want to take advantage of this since it's supported by

by the library already and we will have this this feature also two hours right

so to do this i shared the same document on google drive

so we can see it here so this is the document right with the

same data i don't know if you can see it but it's

the same data and i said this is a public file right

um of course if this would be private we

need to use the google api it will complicate things a lot

so for now we'll just assume that this document is public

for the sake of example but it's still still a useful feature to have even

if it only works with public documents because there's a lot of this kind of

files from governments from institutions that are

public they don't need to be hidden and it's still useful for us to to graph

them right so let's check the same url we can add

it here as the input file right so it could be

this and we just pass it like this

and then we can see if this if this works

um then if it doesn't work uh basically we need to work a bit on this

because there's a chance that this function

raises an exception right so we're going to need to capture that

but first let's try it and see if it works

so i can do like make run or before this let's create a new branch

so basically um yeah let's undo this really quick

okay this do it

and create any brands let's check out and the branch is going to be read from

your drive and now we'll do our changes so the same

we had before it's basically

this is going to be like this with a url and then here we want to deal with the

exception but only after it works so let's just remember here like

exception so we don't forget x option and then there's also

here type spread okay maybe i forgot all the typos but

i'm trying to have it correct and let's try this and

see and see if it works

so we do a make one and file format not supported okay

so after checking a bit it seems like this is like a

an issue with this format so basically we will drive

just like different ways to link to link to the to the url

to link to the file so one of them is this like sharing thing that you get

when you try to share the document but there's other formats that

work so i found i found one that seems to work so

we will use that basically it's instead of

this address with a sharing it's an address that looks like this

let me put it out here exactly and then um

basically it will be this and the id of the document which

is all of this part and in this way it's a file that can be downloaded and it's

treated not as a website but as a file so to do this let's let's see first we

need to check if this is a google doc file and then we will do this

so this google docs by default this has to be

false but we can just check basically the input file

this is uh as we saw this is all of this or the

csv so we don't want it to trigger when it's csp right

or excel or anything else and so let's see

how can we detect that this is a google drive link so

so that the user doesn't have to specify anything no it's all automatic

so i would say when the input file has like http so let's say it starts

with http this covers hdb and cps right

because both start with http and also that it contains like

drive.google.com so i would say this is already like a

good way to detect that this is a google docs

and then if it is in data google docs um

then we can do this trick of changing the address right

so for this we're gonna overwrite the input file

variable which contains the url and we're gonna get uh basically we're

gonna extract the the id right

and to do that uh this is like this small trick

we split it per slash so with this we would have each of these

lashes splitting this into a list and then we

want from the end this would be like minus one from the

end and the id is minus two from the end so just like this in python it's very

easy we just do minus two and this input file will have the id

yes so since this has the id no the id would be this plus

okay exactly so we want all this url plus input file split like this

exactly so now we'll have all of this and then after the id equals and all of

this extract the id actually we can just extract this id if

it's clear let's say document id dot id

and then it's a bit clearer like this okay we can concatenate like this

like this much i think it's clearer even like maybe just like

[Music] long

okay so now this input file will contain the right link

and maybe this now will be able to be downloaded so

hopefully then it matches into the into the text style or the read csv and

it will download it we just need to deal with the exception

here and a couple of features but for now let's add this and then

the good thing about this is that the user doesn't need to know about this uh

special url that google requires so they can just

pass here the the sharing url that you get when you

when you just click here like share well i'm not logged in but uh

when you click here share you just get this with each other if you copy paste

this then that that's what you what you could

use here so let's just try that and see if it

works and then we'll deal with this exception here

still not supported so let's print to see what what exactly we are we are

having here so uh this input file right is overdose

http drive.google.com exactly so let's just print this

should enter here okay

export download an id and the id okay and the opposite set is not

supported and it says it's not supported because

it's entering this point um so i would say maybe instead of

having all these options we're gonna have a different approach

which is like we just try to open it if it's not a csv

we're gonna try to open it and then this gives us an opportunity to deal with

this actually so we try to open it

and then if it doesn't work we will capture the exception so

basically we can do it like this accept let's say that's an exception

as e and then we're going to pin the error here instead of

like before opening file these tricks with the printing

and then we print the exception and then we exit let's try like this

this way um basically if it's a csv we'll try

to read the csv and if not we'll try to read

anything that the retex does and instead of

exploding if there's an exception we just capture it and print this nice

message and then maybe we don't need to print

the exception but for now let's pull it just to see if it works

so we do make run now it's taking a bit longer that looks

good so let's see if it outputted

so this should be the file line jpg so here it is

so we downloaded the file from the internet and it rendered it in a graph

then this works we can remove then this exception maybe we can we can leave it

uh it shouldn't be often that this happens

so maybe it's a good uh debugging thing yeah but basically we

can leave it then whatever this read excel takes our

program takes and that's the small trick with the with

the google link also we could add now that we are doing

some error messaging and we can add something to the size

option because we are not checking actually

we don't have any that nice error here so i could say this is an additional

error handling that i saw right now because uh remember

here we have a size that looks like this x 500 or any two numbers that could be

that the user passes like like 700 500

maybe so we just want to make sure that this is like friendly to the user so

maybe we can say if like if there's no x

in an output size that will print a nice message for for

the user uh wrong size format

we can say like be

something like this in case it tells the user no

then we need an example uh yeah so here this is like another

like friendly message so the more of this

kind of message that we have the the more usable our program will be now even

though it's in the command line it doesn't mean that

there's no user experience so we can adjust this

this style of messages i think this covers

pretty well the the reading from url functionality and this

and this extra messaging here that we added so let's go back

um let's first verify that it still works after our change

always a good idea and it seems to work and now we can just commit it so if we

see the div and we can see our changes yeah not a

lot of changes but we can commit them

and this will be like click excel also takes url

from google drive now and here we can have the minor

changes in error reporting and then

okay so this is a pretty good summary i think the features we implemented

and then we just go and merge it again so let's go to master

and merge this one that is called with google drive here it is without its

tattoos and we can push