BaseHttpServer

[Paul]

Hi,
I currently have a webserver using BaseHttpServe that serves images
like this:
if self.path.endswith(".jpg"):
print(curdir + sep + self.path)
f = open(curdir + sep + self.path,"b")
self.send_response(200)
self.send_header('Content-type', 'image/jpg')
self.end_headers()
self.wfile.write(f.read())
f.close()
return
Whilst it works, it does take quite a while to load (approx 10secs for
a 4mb file even though its over the local connection) - does anyone
have any hints/tips for speeding it up?
Thanks,
Paul

 

[Pierre Quentel]

Hi,
I currently have a webserver using BaseHttpServe that serves images
like this:
if self.path.endswith(".jpg"):
                print(curdir + sep + self.path)
                f = open(curdir + sep + self.path,"b")
                self.send_response(200)
                self.send_header('Content-type',        'image/jpg')
                self.end_headers()
                self.wfile.write(f.read())
                f.close()
                return
Whilst it works, it does take quite a while to load (approx 10secs for
a 4mb file even though its over the local connection) - does anyone
have any hints/tips for speeding it up?
Thanks,
Paul

Hi,

This is probably because you first load the file content in memory by
f.read() before sending it. In this case it's better to send the
content by chunk, using shutil : replace

self.wfile.write(f.read())
by :
shutil.copyfileobj(f,self.wfile)

- Pierre

 

[Steve Holden]

Hi,
I currently have a webserver using BaseHttpServe that serves images
like this:
if self.path.endswith(".jpg"):
print(curdir + sep + self.path)
f = open(curdir + sep + self.path,"b")
self.send_response(200)
self.send_header('Content-type', 'image/jpg')
self.end_headers()
self.wfile.write(f.read())
f.close()
return
Whilst it works, it does take quite a while to load (approx 10secs for
a 4mb file even though its over the local connection) - does anyone
have any hints/tips for speeding it up?

You could consider reading the file in smaller blocks and writing the
output in a loop. That way the next block of the file can be read in
while the network buffers are emptying.

Just keep reading data and writing it until the number of data bytes you
wrote is fewer than the number you tried to read.

regards
Steve

 

[Paul]

You could consider reading the file in smaller blocks and writing the
output in a loop. That way the next block of the file can be read in
while the network buffers are emptying.

Just keep reading data and writing it until the number of data bytes you
wrote is fewer than the number you tried to read.

regards
 Steve

Thanks for these but they seem to have made no difference - it still
loads (visually) in chunks on the screen taking quite a while.
Intrestingly, with the looping idea I got it to print out on each loop
and some took a substantial amount of time to load compared to others
(I tried using 1500 and then 10,000 size chunks,with the code being:
if self.path.endswith(".jpg"):
print(curdir + sep + self.path)
f = open(curdir + sep + self.path,"rb")
self.send_response(200)
self.send_header('Content-type', 'image/jpg')
self.end_headers()
h = 10000
while h==10000:
g = f.read(10000)
h = len(g)
print h
self.wfile.write(g)

#self.wfile.write(f.read())

#shutil.copyfileobj(f,self.wfile)
f.close()
return

I also tried loading it to memory before it was requested but that
also made no difference.
If anyone's got any suggestions, I would be very greatful,
Paul