在python中解析文件以首先找到一个字符串，然后解析以下字符串，直到找到另一个字符串

Question

I am trying to scroll trough a result file that one of our process print.

The objective is to look through various blocks and find a specific parameter. I tried to tackle this but can't find an efficient way that would avoid to parse the file multiple times.

This is an example of the output file that I read:

ID:13123
Compound:xyz
... various parameters
RhPhase:abc

ID:543
Compound:lbm
... various parameters

ID:232355
Compound:dfs
... various parameters
RhPhase:cvb

I am looking for a specific ID that has a RhPhase in it, but since the file contains many more entry, I just want that specific ID. It may or may not have an RhPhase in it; if it has one, I get the value.

The only way that I figured out is to actually go through the whole file (which may be hundreds of blocks, to give an idea of the size), and make a list for each ID that has a RhPhase, then in second instance, I scroll through the dictionary, retrieving the value for a specific ID.

This feels so inefficient; I tried to do something different, but got stuck at how you mark the lines while you scroll through them; so I can tell python to read each line->when find the ID that I want continue to read->if you find RhPhase get the value, otherwise stop at the next ID.

I am stuck here:

datafile=open("datafile.txt", "r")
for items in datafile.readline():
    if "ID:543" in items:
        [read more lines]
        [if "RhPhase" in lines:]
        [    rhphase=lines     ]
        [elif ""ID:" in lines  ]
        [    rhphase=None      ]
        [    break             ]

Once I find the ID; I don't know how to continue to either look for the RhPhase string or find the first ID: string and stop everything (because this means that the ID does not have an associated RhPhase).

This would pass through the file once, and just check for the specific ID, instead of parse the whole thing once and then do a second pass. Is possible to do so or am I stuck to the double parsing ?

Answer 1

Usually, you solve these kind of things with a simple state machine: You read the lines until you find your id; then you put your reader into a special state that then checks for the parameter you want to extract. In your case, you only have two states: ID not found, and ID found, so a simple boolean is enough:

foundId = False
with open('datafile.txt', 'r') as datafile:
    for line in datafile:
        if foundId:
            if line.startswith('RhPhase'):
                print('Found RhPhase for ID 543:')
                print(line)

                # end reading the file
                break
            elif line.startswith('ID:'):
                print('Error: Found another ID without finding RhPhase first')
                break

        # if we haven’t found the ID yet, keep looking for it
        elif line.startswith('ID:543'):
                foundId = True