I want regex for the given sentence to pull just character till 332.5 million units starting from Recalled units

4 StericycleExpertSolutions.com Pharmaceutical recalls increased 52% to 105 – the highest since Q3 2013 and the third highest since at least 2005. Recalled units jumped more than 100 times to nearly 332.5 million units – the highest quarter by far since Q2 2007 and second highest since 2005. This is mostly due to one large recall. CONNECTING THE DOTS 1. Failed specifications made up the top cause of recalls for the seventh consecutive quarter. 2. 84 companies reported at least one recall in Q1 2018 – the highest number since at least 2012. 3. Class I recalled units increased for the second quarter in a row – this time by 192.8%. There were more than 1 million Class I recalled units in Q1 2018 – higher than any quarter from 2016 and three of the quarters in 2017. OF PHARMACEUTICAL RECALLS WERE NATIONWIDE % OF PHARMACEUTICAL UNITS BY CLASS PHARMACEUTICAL RECALLS & UNITS 79% TOP PHARMACEUTICAL CAUSES BASED ON RECALLS Largely due to one large recall, the top cause based on units was Current Good Manufacturing Practice (CGMP) deviations, at 94.7% of recalled units. OUT OF PRACTICE Failed Specs 21.9% Sterility 15.2% Mislabeling 14.3% Foreign Materials 11.4% Subpotency 9.5% Class II 97.8% Class III 1.3% Unclassified 0.6% Class I 0.3% 60 80 140 20 40 120 100 150MM 100MM 50MM 250MM 200MM 350MM 300MM Q1 2017 Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q1 2016 Q2 2016 Q3 2016 Q4 2016

I am using this regexp given below but not able to resolve

recall_data=regexp(pdf_text,(':?Recalled units .*? to \w* [0-9]*+\.*[0-9]*||:?[0-9]*% to ?\w*+\s*+\w* [0-9]*+\.*[0-9]* ?\w*'),'match')

Answer 1

[Cris LaPierre]

It seems to work, though it returns 2 matches. I think the second one if the one you want ?

recall_data = 
    {'52% to 105 '}    {'Recalled units jumped more than 100 times  to nearly 332.5'}

If you just want the 2nd match, your regexp can be simplified to the following

recall_data=regexp(pdf_text,':?Recalled units .*? to \w* [0-9]+\.*[0-9]','match')

[Alfaz Memon]

working thanks. Can you help with the updated data.

[madhan ravi]

Then please be kind enough to accept his answer.

[Alfaz Memon]

please help me with updated queston.

[Cris LaPierre]

You're creating a moving target...

You've updated the text but not the question. The code I provided earlier still answers your question - it gives you the text from 'recalled' to '332.5'.

Could you be more specific on what you want it to do? Easiest is if you show us what your expected output is and what code you've tried, along with a description of why its output is incorrect.

Answer 2

[dfaeaser aseraweraw]

GREAT