#To start:
$ cqp -e

#To see available corpora:
 show corpora

#Show information about a given corpus:
 info BNC-XML

#Activate corpus:
 BNC-XML

#Show corpus attributes:
 show cd

#Search for words (via regular expressions) and sort:
 "spe(ech|aks?(ing)?)"
 sort by word

#Set context to 8 words preceding the target, 2 sentences following the target, 1 sentence pre- and post-target:
 set lc 8 words
 set rc 2s
 set c s

#Redisplay matches:
 cat

#Display or hide POS and lemma annotation:
 show +pos
 show +lemma
 show -pos -lemma

#Search by lemma:
 [lemma = "speak_VERB"]
 [lemma = "speech_SUBST"]
 [lemma = "(speak_VERB|speech_SUBST)"]

#See size of last query:
 size Last

#Show structural attributes (shown as XML tags):
 show +s

#Create .cqprc file with favorite settings:
set ProgressBar on;
set HistoryFile "/tmp/cqphistory.jdegen";
set WriteHistory yes;
set c s;

#Searching for POS information:
 "work"
 [word="work" & pos="N.*"]
 [word="work" & pos="V.*"]
 [word="work" & pos !="V.*"]

#Use \verb|/codist[]| macro to get frequency distributions of POS-tags/lemmas over a given word:
 /codist["work",  pos]
 /codist[lemma, "speak_VERB", word]


#Search for sequences, search within a context:
 [lemma="work_VERB"][]*[word="day"]
 [lemma="work_VERB"][]*[word="day"] within s
 [lemma="work_VERB"][]*[word="day"] within 2 words
 [lemma="work_VERB"][]{2}[word="day"]

#Count: 
 count by word
 count by lemma

#Set frequency thresholds:
 [pos="VVB" & word = "w.*"]
 count by lemma cut 50

#Save query results:
 Some = [word = "some" %c] [pos="NN2*"]
 set DataDirectory "."
 BNC-XML
 save Some
 cat Some > "some.txt"
 cat Some > "| gzip > some.txt.gz"
 sort Some by word

#Anchor points:
 A = [pos="(AT.*|DT.*)"] @[pos="AJ.*" & word="f.*"] [pos="N.*"]
 sort by word

#Display corpus positions of anchor points in tabular format:
 dump A
 dump A 10 20

#Frequency distributions:
 group A matchend word by target word cut 100
 group A match word by target lemma cut 100

#Reduce data randomly:
 reduce A to 10%