# W A Lorenz 10-2009
# quick class that is useful for parsing various
# faa like formats from a file.

from subprocess import Popen, PIPE

class parseFaa:
    def __init__(self, fileName):
        self.fileName = fileName
        self.file = open(fileName, 'r')
        self.done = False
    def next(self):
        if self.done==True:
            return ('','')
        seq=''
        name=''
        done=False
        while(done==False):
            line=self.file.readline()
            if not line:
                self.file.close()
                self.done=True
                done=True
            elif line[0]=='>' and len(line.split())>1:
                name=line.split()[1]
            elif line[0].upper() in ['A','C','G','U']:
                seq=line.split()[0]
                done=True
        return (seq, name)
    def nextSeqStrMFE(self):
        if self.done==True:
            return ('','', '', '')
        seq=''
        name=''
        struc=''
        strucMFE=''
        done=False
        seqDone=False
        strucDone=False
        while(done==False):
            line=self.file.readline()
            if not line:
                self.file.close()
                self.done=True
                done=True
                continue
            elif line[0]=='>' and len(line.split())>1:
                name=line.split()[1]
            elif line[0] in ['.','(',')']:
                struc=line.split()[0]
                strucDone=True
            elif line[0].upper() in ['A','C','G','U']:
                seq=line.split()[0]
                seqDone=True
            done=seqDone and strucDone
        if self.done==True:
            return ('','','','')
        p=Popen("RNAfold", stdin=PIPE, stdout=PIPE)
        (pout, perr)=p.communicate(seq+'\n')
        strucMFE=pout.split('\n')[1].split()[0]
        return (seq, name, struc, strucMFE)
    def nextSeq2Str(self):
        if self.done==True:
            return ('','', '', '')
        seq=''
        name=''
        struc=''
        strucMFE=''
        done=False
        seqDone=False
        strucDone=False
        struc2Done=False
        while(done==False):
            line=self.file.readline()
            if not line:
                self.file.close()
                self.done=True
                done=True
                continue
            elif line[0]=='>' and len(line.split())>1:
                name=line.split()[1]
            elif line[0] in ['.','(',')'] and not strucDone:
                struc=line.split()[0]
                strucDone=True
            elif line[0] in ['.','(',')'] and strucDone:
                struc2=line.split()[0]
                struc2Done=True
            elif line[0].upper() in ['A','C','G','U']:
                seq=line.split()[0]
                seqDone=True
            done=seqDone and struc2Done
        if self.done==True:
            return ('','','','')
        return (seq, name, struc, struc2)
    def nextall(self):
        if self.done==True:
            return ('','','','')
        seq='';name='';seqline='';nameline=''
        done=False
        while(done==False):
            line=self.file.readline()
            if not line:
                self.file.close()
                self.done=True
                done=True
            elif line[0]=='>' and len(line.split())>1:
                name=line.split()[1]
                nameline=line
            elif line[0].upper() in ['A','C','G','U']:
                seq=line.split()[0]
                seqline=line
                done=True
        return (seq, name, seqline, nameline)
    def close(self):
        self.file.close()