将一个函数列表应用于列表中的每个项目？

Question

I am making a program that works on chat files under the following general form:

1. Open file & import lines (via readlines )
2. Do an initial pass to turn the list of strings into better-formed data (currently dicts) by:
   • Throwing out out malformed lines
   • Separating out the usernames from the text of the message
   • Throwing out lines for ignored (typically bot) users
   • Marking which lines are /me commands
3. Do a second pass over the list of dicts to apply various manipulations on them, such as:
   • Replacing every mention of a nick with its alias
   • Applying special formatting to /me commands

Rather than have multiple switches in the config file and lots of if statement checks within the loops, I believe the program would be cleaner if I generated the list of functions elsewhere and then fed the program the list of dicts (or strings, depending on which part of the program I'm at) and a list of functions such that the list of functions gets applied to each of the items in the list of objects.

It seems that this probably would be a good case for list comprehensions if I were only applying a single function to each item, but I don't want to have to do a separate pass through the log for every function that I want to call. However, this answer notes that list comprehension probably aren't what I want, since they return a completely new list , rather than modifying in place.

---

Is my best option to have two variants of the following?

for item in list:
    item = a(b(c(d(item, dparam1, dparam2), cparam)), aparams)

(for readability, I'd put each function on its own line, like:

for item in list:
    item = d(item, dparam1, dparam2)
    item = c(item, cparam)
    item = b(item)
    item = a(item, aparams)

However, the above doesn't eliminate the need for if checks on all the switches and wouldn't allow for applying function a at two different places unless I explicitly add a switch to do so.

Answer 1

Based on your sample code, you may try this way,

func_list = [d, c, b, a]
for idx, item in enumerate(item_list):
    item_list[idx] = reduce(lambda v, func: func(v), func_list, item)

Let each func handle the data, make the loop more like pipeline.